diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000000000000000000000000000000000000..a6344aac8c09253b3b630fb776ae94478aa0275b --- /dev/null +++ b/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000000000000000000000000000000000000..d4c9b9c7d794c5cf874a55b1cb34d91eb8ba1b20 --- /dev/null +++ b/LICENSE @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2026 ACappellaSet contributors + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..a0ceae90245398b4897697f2d17beaa297988085 --- /dev/null +++ b/README.md @@ -0,0 +1,100 @@ +--- +license: cc-by-nc-4.0 +tags: +- audio +- music-source-separation +- source-separation +- htdemucs +- acappella +inference: false +--- + +# ACappellaSet models + +Model weights and release metadata for a cappella source separation. + +**Repository status: private.** The active release contains **29 checkpoints**: three fixed-output models, twelve original part-specific specialists, and fourteen conditioned selections, including two trained with AI-generated singer augmentation. Downloads require repository access. +**Weight license: [CC BY-NC 4.0](licenses/CC-BY-NC-4.0.txt)**, subject to the upstream notices and third-party rights described in [WEIGHTS_NOTICE.md](WEIGHTS_NOTICE.md). + +- **Checkpoint paths, hashes, configurations, and validation selections:** [manifest.json](manifest.json) +- **Code and installation:** [ACappellaSet on GitHub](https://github.com/pting126/ACappellaSet) +- **Inference:** [fixed-output and part-specific models](https://github.com/pting126/ACappellaSet/blob/main/docs/inference.md), [shared query-conditioned models](https://github.com/pting126/ACappellaSet/blob/main/docs/conditioned-inference.md). +- **Training and configurations:** [published code and recipes](https://github.com/pting126/ACappellaSet). +- **Dataset:** [ACappellaSet](https://huggingface.co/datasets/AcappellaSet/ACappellaSet) is released with Manual Approval for Non-commercial Research Use. Dataset access approval does not grant access to this private model repository, and Dataset Terms do not relicense these model weights. + +## Fixed-output and part-specific models + +These models use the [HTDemucs-MSS architecture](https://github.com/pting126/ACappellaSet/blob/main/models/demucs4ht.py), with parallel waveform and spectrogram branches coupled by a cross-domain Transformer. Fixed-output models predict six vocal parts simultaneously. Each part-specific model has a two-output head for its designated target and residual, with a separate checkpoint per vocal part; no query input is required. + +JC denotes JaCappella; SK denotes natural Seagull-K recordings; JC + SK denotes joint training. Each checkpoint below links to its file. + +| Model | Training corpus | Checkpoints | +| --- | --- | --- | +| Fixed-output HTDemucs (6 outputs) | JC | [Fixed_6stem_htdemucs_JC](fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt) | +| Fixed-output HTDemucs (6 outputs) | SK | [Fixed_6stem_htdemucs_SK](fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt) | +| Fixed-output HTDemucs (6 outputs) | JC + SK | [Fixed_6stem_htdemucs_SKnJC](fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt) | +| Part-specific HTDemucs (mixture input) | JC | [Soprano](specialists/JC/legacy/soprano/epoch_0162.pt), [Alto](specialists/JC/legacy/alto/epoch_0223.pt), [Tenor](specialists/JC/legacy/tenor/epoch_0119.pt), [Bass](specialists/JC/legacy/bass/epoch_0110.pt), [VP](specialists/JC/legacy/vp/epoch_0045.pt), [Lead](specialists/JC/legacy/lead/epoch_0149.pt) | +| Part-specific HTDemucs (mixture input) | SK | [Soprano](specialists/SK/legacy/soprano/epoch_0220.pt), [Alto](specialists/SK/legacy/alto/epoch_0237.pt), [Tenor](specialists/SK/legacy/tenor/epoch_0214.pt), [Bass](specialists/SK/legacy/bass/epoch_0182.pt), [VP](specialists/SK/legacy/vp/epoch_0205.pt), [Baritone](specialists/SK/legacy/baritone/epoch_0043.pt) | + +## Shared query-conditioned + +These models extend [HTDemucs with query-conditioned FiLM layers](https://github.com/pting126/ACappellaSet/blob/main/models/conditioned_core.py) in the waveform and spectrogram branches before the cross-domain Transformer. Queries use a learned vocal-part embedding, a matching fine-tuned ReDimNet2 singer-enrollment encoder, or both. A shared model extracts the requested target and its residual; the full-remix and single-clone models use the singer-conditioned architecture with different training data. + +Checkpoint labels show the selected training step and validation criterion. “Pooled” means JC + SK validation. + +| Model | Training data | Checkpoint (validation criterion) | +| --- | --- | --- | +| Vocal-part-conditioned HTDemucs | SK | [4k](conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt) (JC inclusive); [198k](conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt) (SK shared) | +| Vocal-part-conditioned HTDemucs | JC | [119k](conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt) (SK inclusive); [141k](conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt) (JC shared) | +| Singer-conditioned HTDemucs | SK | [12k](conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt) (JC shared, JC inclusive); [106k](conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt) (pooled shared, pooled inclusive); [179k](conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt) (SK shared) | +| Vocal-part + singer-conditioned HTDemucs | SK | [7k](conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt) (JC inclusive) | +| Singer-conditioned HTDemucs | JC | [184k](conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt) (SK inclusive) | +| Vocal-part + singer-conditioned HTDemucs | JC | [188k](conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt) (SK inclusive) | +| Singer-conditioned HTDemucs | JC + SK | [195k](conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt) (JC shared, pooled shared); [198k](conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt) (SK shared) | +| Singer-conditioned HTDemucs | SK + full-remix singer augmentation | [131k](conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt) (pooled inclusive) | +| Singer-conditioned HTDemucs | SK + single-clone singer augmentation | [48k](conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt) (pooled inclusive) | + +Conditioned runs were trained through 200k updates; the released files were selected using validation scores. Training data and validation corpus are separate attributes. + +Shared aggregates retained S/A/T/B/VP pairs rather than averaging five part means. Inclusive additionally includes JC Lead or SK Bar. In the machine-readable registry, `combined_*` selection labels mean pooled JC + SK validation. + +## Download and inference + +```bash +hf auth login +hf download AcappellaSet/ACappellaSet-Models --local-dir ./ACappellaSet-Models +``` + +To download one checkpoint, add its file path from the tables above after the repository name. Follow the linked GitHub installation and inference guides and pass the downloaded file with `--checkpoint`. The registry records `inference_model_id` and, for shared query-conditioned models, the available `selection_labels`. Released files contain inference tensors only (plus the selected step for query-conditioned models) and load with `weights_only=True`. Checkpoint and configuration hashes are verified before strict loading. + +Fixed-output and part-specific models take mixture audio. Shared query-conditioned models additionally use a vocal-part query, singer-enrollment audio, or both, according to their conditioning type. All 29 checkpoints have registered inference adapters. Singer- and dual-conditioned exports include their matching fine-tuned encoder tensors. + +### Output conventions + +The fixed-output JC + SK model returns `vp`, `bass`, `soprano`, `alto`, `tenor`, and `other`. +Its `other` output has JC Lead / SK Baritone training semantics; these are not +independent outputs. Part-specific specialists use residual index 0 and target +index 1. Shared query-conditioned models use target index 0 and residual index 1. Preserve each checkpoint's native source order and inference settings. + +## Layout + +```text +manifest.json +fixed/{JC,SK,SKnJC}/legacy/ +specialists/{JC,SK}/legacy/{part}/ +conditioned/ +configs/ +``` + +`model.json` and conditioned step JSON files record checkpoint identity, configuration, and license. `bank.json` maps specialist roles to model IDs. The registry provides released-file SHA-256 hashes. Configurations are included under `configs/`; maintained executable code is on GitHub. + +Validation-selected models record the validation corpus, shared/inclusive scope, training cutoff, and selected step. Multiple selectors may refer to the same file. The two AI-augmented entries identify their augmentation datasets explicitly. + +## Licenses and citation + +Original repository documentation and metadata use the [MIT license](LICENSE). +All 29 model checkpoints use [CC BY-NC 4.0](licenses/CC-BY-NC-4.0.txt). Preserve attribution and indicate changes when sharing. This project license applies only to rights the licensors can grant; upstream notices and third-party rights remain applicable. See [weight notices](WEIGHTS_NOTICE.md). Dataset audio is separate. +Upstream architecture attribution is maintained in the GitHub repository's +[third-party notices](https://github.com/pting126/ACappellaSet/blob/main/THIRD_PARTY_NOTICES.md). + +Cite the accompanying paper when a verified citation becomes available; do not infer publication status from this repository. diff --git a/WEIGHTS_NOTICE.md b/WEIGHTS_NOTICE.md new file mode 100644 index 0000000000000000000000000000000000000000..ca57f41d6e54d223690d429c7797bc07ab67b260 --- /dev/null +++ b/WEIGHTS_NOTICE.md @@ -0,0 +1,18 @@ +# Model weight licensing status + +**Project weight license: [Creative Commons Attribution-NonCommercial 4.0 International](licenses/CC-BY-NC-4.0.txt) (CC BY-NC 4.0).** This applies to all 29 project checkpoints identified in [manifest.json](manifest.json), only to rights the licensors are authorized to grant. Credit ACappellaSet and preserve upstream attribution; indicate modifications when sharing. It does not grant additional third-party rights or change upstream license terms. The full-remix and single-clone singer-augmentation checkpoints were trained with voice-converted singer augmentation generated using Moises; their project license does not grant additional rights controlled by service providers or other third parties. + +## Upstream notices + +Demucs source code uses MIT; preserve the [Meta notice](licenses/Demucs-MIT.txt). +Pretrained initialization identities and tensor correspondence are recorded in the [initialization provenance](https://github.com/pting126/ACappellaSet/blob/main/third_party/initialization.json). + +Historical fixed and specialist training uses Music-Source-Separation-Training; +preserve its [MIT notice](licenses/MSS-MIT.txt). +Singer- and dual-conditioned models include fine-tuned WeSpeaker ReDimNet2 +B6-LM encoder weights; preserve the upstream [Apache license](licenses/WeSpeaker-Apache-2.0.txt) +and [WeSpeaker attribution](https://huggingface.co/Wespeaker/wespeaker-voxceleb-redimnet2-B6-LM). +Project fine-tuning changes pretrained tensors. All released checkpoints are inference-only exports, retaining every separator and matching encoder tensor while omitting optimizer state, training arguments, and internal filesystem paths. + +Original documentation and metadata retain their [MIT license](LICENSE). +Dataset audio, corpus terms, and third-party rights remain separate. diff --git a/conditioned/README.md b/conditioned/README.md new file mode 100644 index 0000000000000000000000000000000000000000..f45fb3590681631d7f861424a42c6ab8b079d676 --- /dev/null +++ b/conditioned/README.md @@ -0,0 +1,9 @@ +# Shared query-conditioned models + +The release contains **14 selected checkpoints from 9 training runs**, including SK full-remix singer augmentation (131k) and SK single-clone singer augmentation (48k). Both augmentation models use pooled-inclusive validation selection. Single-clone refers to the augmentation dataset, not disabled waveform augmentation. + +See the [model card](../README.md#shared-query-conditioned) and [manifest](../manifest.json) for conditioning modes, training data, selected steps, validation corpora/scopes, and checkpoint hashes. + +[Public inference adapters](https://github.com/pting126/ACappellaSet/blob/main/docs/conditioned-inference.md) support all 14 selections and nine model recipes. Checkpoints include the matching fine-tuned encoder parameters for singer and dual models. + +Released files contain only the selected step and original model/encoder tensors; they load with `weights_only=True`. Preserve the registered enrollment-audio preprocessing. All 14 conditioned weights use CC BY-NC 4.0; see [weight notices](../WEIGHTS_NOTICE.md). diff --git a/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.json b/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.json new file mode 100644 index 0000000000000000000000000000000000000000..889fc5d05cae95c9513735579f91dfec852b6b6e --- /dev/null +++ b/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.json @@ -0,0 +1,38 @@ +{ + "id": "l4-step0188000", + "family": "conditioned", + "research_id": "L4", + "description": "HTDemucs | vocal-part + singer-enrollment-conditioned | jc-natural | redimnet2-finetuned | step 188000 (L4)", + "conditioning": "dual", + "training_data": "jc-natural", + "encoder": "redimnet2-finetuned", + "step": 188000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt", + "original_checkpoint_sha256": "724af8d2906fc08de54b7484df69143a3a7b2c62c412e6e473b979bc34cd74d0", + "original_checkpoint_filename": "best_step0188000.pt", + "released_checkpoint_sha256": "511dc6729da436c5fe72960e9a1de4968578fbb985fd00db5eb653e2d605dfb9", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L4.json", + "config_sha256": "7c46503ca0ddc32623e184e801fe411184ff3a3d42dcddaf34106fb33d45f2cf", + "inference_model_id": "Dual_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161978465, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt b/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt new file mode 100644 index 0000000000000000000000000000000000000000..74209f1890d1597b5ec03d666ad31c8972b5367b --- /dev/null +++ b/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:511dc6729da436c5fe72960e9a1de4968578fbb985fd00db5eb653e2d605dfb9 +size 161978465 diff --git a/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.json b/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.json new file mode 100644 index 0000000000000000000000000000000000000000..dc76b216e821043d716dec891776e7fdb41f1e46 --- /dev/null +++ b/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.json @@ -0,0 +1,38 @@ +{ + "id": "k7-step0007000", + "family": "conditioned", + "research_id": "K7", + "description": "HTDemucs | vocal-part + singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 7000 (K7)", + "conditioning": "dual", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 7000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt", + "original_checkpoint_sha256": "0646527189b1221552c01c918cff01173e98077dc515a6fe9f795ffab00c24ee", + "original_checkpoint_filename": "best_step0007000.pt", + "released_checkpoint_sha256": "528d75380b58ef43e266a0c1f6f3cd86883710d2a95ff9c492724790193fa6ab", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K7.json", + "config_sha256": "5cbc1736ca73762cd34603e55cc3850e2cbf2872388898f710a4ff3e66b117dc", + "inference_model_id": "Dual_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161978465, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt b/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt new file mode 100644 index 0000000000000000000000000000000000000000..df15c6ec79297d3082e7594fc49c53665d7974c6 --- /dev/null +++ b/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:528d75380b58ef43e266a0c1f6f3cd86883710d2a95ff9c492724790193fa6ab +size 161978465 diff --git a/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.json b/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.json new file mode 100644 index 0000000000000000000000000000000000000000..f6a002630b313a3e75ec389cb77b4eb4dc5b110a --- /dev/null +++ b/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.json @@ -0,0 +1,38 @@ +{ + "id": "l3-step0184000", + "family": "conditioned", + "research_id": "L3", + "description": "HTDemucs | singer-enrollment-conditioned | jc-natural | redimnet2-finetuned | step 184000 (L3)", + "conditioning": "singer", + "training_data": "jc-natural", + "encoder": "redimnet2-finetuned", + "step": 184000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt", + "original_checkpoint_sha256": "9f49b7d59575658dcfb9667395d998e8fc34a935b3995b099593e99ee1de5226", + "original_checkpoint_filename": "best_step0184000.pt", + "released_checkpoint_sha256": "7021317a39953018ea31d9f07578bbe4da9e3646ca62e1e59891abba57edd94d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L3.json", + "config_sha256": "a0bab6c22d17b9b23eca07c5cd3f2f41ffcb01d807ea78a19f3c131eccd91438", + "inference_model_id": "Singer_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt b/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt new file mode 100644 index 0000000000000000000000000000000000000000..d8b6cae56a6bab7cf25bcaa3cf0f0d0599d8a177 --- /dev/null +++ b/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7021317a39953018ea31d9f07578bbe4da9e3646ca62e1e59891abba57edd94d +size 161878741 diff --git a/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.json b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.json new file mode 100644 index 0000000000000000000000000000000000000000..6aee38e4dd581ac44ce1be54c69150c0ec2ade37 --- /dev/null +++ b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.json @@ -0,0 +1,44 @@ +{ + "id": "m3-step0195000", + "family": "conditioned", + "research_id": "M3", + "description": "HTDemucs | singer-enrollment-conditioned | jc-sk-natural | redimnet2-finetuned | step 195000 (M3)", + "conditioning": "singer", + "training_data": "jc-sk-natural", + "encoder": "redimnet2-finetuned", + "step": 195000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_shared", + "combined_shared" + ], + "paper_tables": [ + "development_a" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt", + "original_checkpoint_sha256": "badc60d5bb2f2cf0cca92b11dc98c5a4de51fcd41312fff43607f7664ba5f9df", + "original_checkpoint_filename": "best_step0195000.pt", + "released_checkpoint_sha256": "12a92b20a7b1fed9614c48b925eaeb2f563428201d5c270c11e3f7c3df7359e7", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "supersedes": { + "step": 173000, + "sha256": "f99b7ee04a026c2c628bcfaac3ddbe7a5ffabc95bb96e9d0889e3079e8db1b06", + "preserve_historical": true + }, + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/M3.json", + "config_sha256": "fee6def3080111793b362b2b9fda724720ddd13f28b2596f2a3eeb95fb8168d0", + "inference_model_id": "Singer_conditioned_htdemucs_SKnJC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt new file mode 100644 index 0000000000000000000000000000000000000000..ebef7c955064c767b8a51649a9e9c5bc840b9989 --- /dev/null +++ b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12a92b20a7b1fed9614c48b925eaeb2f563428201d5c270c11e3f7c3df7359e7 +size 161878741 diff --git a/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.json b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.json new file mode 100644 index 0000000000000000000000000000000000000000..c0bf794bfd26b2a8eddcaf13a9fa749446254e69 --- /dev/null +++ b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.json @@ -0,0 +1,43 @@ +{ + "id": "m3-step0198000", + "family": "conditioned", + "research_id": "M3", + "description": "HTDemucs | singer-enrollment-conditioned | jc-sk-natural | redimnet2-finetuned | step 198000 (M3)", + "conditioning": "singer", + "training_data": "jc-sk-natural", + "encoder": "redimnet2-finetuned", + "step": 198000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_shared" + ], + "paper_tables": [ + "development_a" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt", + "original_checkpoint_sha256": "6997e980ed2dded55ba1db2d5e88b71ad7024782125b2bd8dcf41df71189d6b4", + "original_checkpoint_filename": "best_step0198000.pt", + "released_checkpoint_sha256": "67c68a7f12076db456166edab0120f94947806689eca50da546f7adac260d77b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "supersedes": { + "step": 159000, + "sha256": "03e74a6ae7f56d647c66e582f368f714b5504f8c657e78fb7f6c2142d0ae8ab1", + "preserve_historical": true + }, + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/M3.json", + "config_sha256": "fee6def3080111793b362b2b9fda724720ddd13f28b2596f2a3eeb95fb8168d0", + "inference_model_id": "Singer_conditioned_htdemucs_SKnJC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt new file mode 100644 index 0000000000000000000000000000000000000000..ab9a1c202e95f1d708ea67b7cf5880deac1e9859 --- /dev/null +++ b/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:67c68a7f12076db456166edab0120f94947806689eca50da546f7adac260d77b +size 161878741 diff --git a/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.json b/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.json new file mode 100644 index 0000000000000000000000000000000000000000..546858057bcd060003e2964ffc92709229f1e9a5 --- /dev/null +++ b/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.json @@ -0,0 +1,38 @@ +{ + "id": "k6-step0131000", + "family": "conditioned", + "research_id": "K6", + "description": "HTDemucs | singer-enrollment-conditioned | sk-full-remix | redimnet2-finetuned | step 131000 (K6)", + "conditioning": "singer", + "training_data": "sk-full-remix", + "encoder": "redimnet2-finetuned", + "step": 131000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "combined_inclusive" + ], + "paper_tables": [ + "development_b" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt", + "original_checkpoint_sha256": "ff74b497341edcfe6a6357749ed3e6bff795f1740ef022f4003f1241f198e517", + "original_checkpoint_filename": "best_step0131000.pt", + "released_checkpoint_sha256": "893326c96ac17b9cfed1cd5126176763f6a0a525115e3322c9a701ef232a9e69", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "artifact_format": "acappellaset-conditioned-inference-v1", + "notice_path": "WEIGHTS_NOTICE.md", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "config": "configs/conditioned/Singer_conditioned_htdemucs_SK_full_remix.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "released_checkpoint_bytes": 161878741, + "inference_model_id": "Singer_conditioned_htdemucs_SK_full_remix", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt b/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt new file mode 100644 index 0000000000000000000000000000000000000000..b7cf4837903b281988a4b5c5b3b81f43c4f60dd4 --- /dev/null +++ b/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:893326c96ac17b9cfed1cd5126176763f6a0a525115e3322c9a701ef232a9e69 +size 161878741 diff --git a/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.json b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.json new file mode 100644 index 0000000000000000000000000000000000000000..d33a262d1d9b7f2ab1b44f07820d4f6b38795d2b --- /dev/null +++ b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.json @@ -0,0 +1,40 @@ +{ + "id": "k5-step0012000", + "family": "conditioned", + "research_id": "K5", + "description": "HTDemucs | singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 12000 (K5)", + "conditioning": "singer", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 12000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_shared", + "jc_inclusive" + ], + "paper_tables": [ + "development_a", + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt", + "original_checkpoint_sha256": "6f8e199008a542904330495dab28d7f56d8c380a6959ea4e460b22450c56ca43", + "original_checkpoint_filename": "best_step0012000.pt", + "released_checkpoint_sha256": "89c5a8829aca66cdbb0033338dc83d769e06f98300f68464a666b87f73e80abf", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K5.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "inference_model_id": "Singer_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt new file mode 100644 index 0000000000000000000000000000000000000000..c11f993a960f9daa1f08a3a92abf661d604e19d7 --- /dev/null +++ b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89c5a8829aca66cdbb0033338dc83d769e06f98300f68464a666b87f73e80abf +size 161878741 diff --git a/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.json b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.json new file mode 100644 index 0000000000000000000000000000000000000000..08c0dfd3b733620f7676e411e5d93cbdfc459e21 --- /dev/null +++ b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.json @@ -0,0 +1,40 @@ +{ + "id": "k5-step0106000", + "family": "conditioned", + "research_id": "K5", + "description": "HTDemucs | singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 106000 (K5)", + "conditioning": "singer", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 106000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "combined_shared", + "combined_inclusive" + ], + "paper_tables": [ + "development_a", + "development_b" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt", + "original_checkpoint_sha256": "df846b039ddd5e032f221adcf3de33a9a3f546f21b47f0bc3d30df891ab4ca93", + "original_checkpoint_filename": "best_step0106000.pt", + "released_checkpoint_sha256": "b28626e30a12ba43337b3cec3da2558685ddcc649fe34dd78e7e71590e922b29", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K5.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "inference_model_id": "Singer_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt new file mode 100644 index 0000000000000000000000000000000000000000..9b49a311acf9e20b9903d857584651a3af565114 --- /dev/null +++ b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b28626e30a12ba43337b3cec3da2558685ddcc649fe34dd78e7e71590e922b29 +size 161878741 diff --git a/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.json b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.json new file mode 100644 index 0000000000000000000000000000000000000000..5d3ae5b5b2e0c338d2a12669d270970dfee84bdb --- /dev/null +++ b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.json @@ -0,0 +1,38 @@ +{ + "id": "k5-step0179000", + "family": "conditioned", + "research_id": "K5", + "description": "HTDemucs | singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 179000 (K5)", + "conditioning": "singer", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 179000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_shared" + ], + "paper_tables": [ + "development_a" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt", + "original_checkpoint_sha256": "eeeac981bacdd063207a0258fb60625edd2b60e94bc544852a1c2459223a04d3", + "original_checkpoint_filename": "best_step0179000.pt", + "released_checkpoint_sha256": "537531081da4c7c1a7d5568f8847bfaab1df3d54658486381bdd7f3d2f7f8d7d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K5.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "inference_model_id": "Singer_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt new file mode 100644 index 0000000000000000000000000000000000000000..091210e988ef692cd77de441fc44be14ca2f4511 --- /dev/null +++ b/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:537531081da4c7c1a7d5568f8847bfaab1df3d54658486381bdd7f3d2f7f8d7d +size 161878741 diff --git a/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.json b/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.json new file mode 100644 index 0000000000000000000000000000000000000000..4e085fffaa8011a356ad3299032f90cd04f99254 --- /dev/null +++ b/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.json @@ -0,0 +1,38 @@ +{ + "id": "k6-no-aug-step0048000", + "family": "conditioned", + "research_id": "K6_no_aug", + "description": "HTDemucs | singer-enrollment-conditioned | sk-single-clone | redimnet2-finetuned | step 48000 (K6_no_aug)", + "conditioning": "singer", + "training_data": "sk-single-clone", + "encoder": "redimnet2-finetuned", + "step": 48000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "combined_inclusive" + ], + "paper_tables": [ + "development_b" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt", + "original_checkpoint_sha256": "79ee3dd239925ea2c075641eadc0e13c001a81f9f367f8ee6881c5e8a3881901", + "original_checkpoint_filename": "best_step0048000.pt", + "released_checkpoint_sha256": "00fc378e77c7148f75e853ba8e84f246aba1557c30ec3e5f205e4227099c5c65", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "artifact_format": "acappellaset-conditioned-inference-v1", + "notice_path": "WEIGHTS_NOTICE.md", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "config": "configs/conditioned/Singer_conditioned_htdemucs_SK_single_clone.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "released_checkpoint_bytes": 161878741, + "inference_model_id": "Singer_conditioned_htdemucs_SK_single_clone", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt b/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt new file mode 100644 index 0000000000000000000000000000000000000000..435f80188d3b282842d36ea0a932400209aa5b51 --- /dev/null +++ b/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:00fc378e77c7148f75e853ba8e84f246aba1557c30ec3e5f205e4227099c5c65 +size 161878741 diff --git a/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.json b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.json new file mode 100644 index 0000000000000000000000000000000000000000..7dce2b6c60a2b938c0b657b0403ff3d9519caabf --- /dev/null +++ b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.json @@ -0,0 +1,38 @@ +{ + "id": "l1-step0119000", + "family": "conditioned", + "research_id": "L1", + "description": "HTDemucs | vocal-part-conditioned | jc-natural | no-encoder | step 119000 (L1)", + "conditioning": "part", + "training_data": "jc-natural", + "encoder": "no-encoder", + "step": 119000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt", + "original_checkpoint_sha256": "5de579abaf11546eab1fcfa29566cf4934ece6c5fe7b23df56b6af27c3be3723", + "original_checkpoint_filename": "best_step0119000.pt", + "released_checkpoint_sha256": "6af03b62d8e5f419f72787b7909c3e82ad7d3194fa43c920a47a064374f63e8b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L1.json", + "config_sha256": "45fca96de4f1849d076368538dd789f97e5e0a4ab4443c6c0922e4505c2ac08a", + "inference_model_id": "Part_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt new file mode 100644 index 0000000000000000000000000000000000000000..533cbfd9d14666a47ee67bbdf0ce2f07f36d281b --- /dev/null +++ b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6af03b62d8e5f419f72787b7909c3e82ad7d3194fa43c920a47a064374f63e8b +size 109889741 diff --git a/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.json b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.json new file mode 100644 index 0000000000000000000000000000000000000000..6f3c7872fc2316b4f59d25d4605a7161974ca50b --- /dev/null +++ b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.json @@ -0,0 +1,38 @@ +{ + "id": "l1-step0141000", + "family": "conditioned", + "research_id": "L1", + "description": "HTDemucs | vocal-part-conditioned | jc-natural | no-encoder | step 141000 (L1)", + "conditioning": "part", + "training_data": "jc-natural", + "encoder": "no-encoder", + "step": 141000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_shared" + ], + "paper_tables": [ + "cross_corpus" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt", + "original_checkpoint_sha256": "92e19968bef2ce84d3800141afc50cc47141bd016adfcc07b70e705df7408004", + "original_checkpoint_filename": "best_step0141000.pt", + "released_checkpoint_sha256": "f9fdca2de1f89b650b603d386f39128b4b06b0ebd4a59469f57ce256e8c26065", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L1.json", + "config_sha256": "45fca96de4f1849d076368538dd789f97e5e0a4ab4443c6c0922e4505c2ac08a", + "inference_model_id": "Part_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt new file mode 100644 index 0000000000000000000000000000000000000000..8cabf1564a83476e193b075fa0bfe24a504b1aa2 --- /dev/null +++ b/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9fdca2de1f89b650b603d386f39128b4b06b0ebd4a59469f57ce256e8c26065 +size 109889741 diff --git a/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.json b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.json new file mode 100644 index 0000000000000000000000000000000000000000..976eeaee409cb6844d5acc2c1de48544d1017906 --- /dev/null +++ b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.json @@ -0,0 +1,38 @@ +{ + "id": "k1-step0004000", + "family": "conditioned", + "research_id": "K1", + "description": "HTDemucs | vocal-part-conditioned | sk-natural | no-encoder | step 4000 (K1)", + "conditioning": "part", + "training_data": "sk-natural", + "encoder": "no-encoder", + "step": 4000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt", + "original_checkpoint_sha256": "979ba17521f398fb1b2a3e0ccad29615999b50e87d76347b2682212a61b49427", + "original_checkpoint_filename": "best_step0004000.pt", + "released_checkpoint_sha256": "31821cfe956fc6a7527e718d6702bcd71b84a8881484c1339753bca7207b316b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K1.json", + "config_sha256": "a2c358f669574dafa25403b5d6e4a0ee0fc3cf237e03399886761f95fd7c0fbd", + "inference_model_id": "Part_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt new file mode 100644 index 0000000000000000000000000000000000000000..7cb545d44dd774cec12ea95a3d91d3c4e4a0f0b7 --- /dev/null +++ b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31821cfe956fc6a7527e718d6702bcd71b84a8881484c1339753bca7207b316b +size 109889741 diff --git a/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.json b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.json new file mode 100644 index 0000000000000000000000000000000000000000..b35362a82e0800f7e13000d2420b87285fc17a3c --- /dev/null +++ b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.json @@ -0,0 +1,43 @@ +{ + "id": "k1-step0198000", + "family": "conditioned", + "research_id": "K1", + "description": "HTDemucs | vocal-part-conditioned | sk-natural | no-encoder | step 198000 (K1)", + "conditioning": "part", + "training_data": "sk-natural", + "encoder": "no-encoder", + "step": 198000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_shared" + ], + "paper_tables": [ + "cross_corpus" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt", + "original_checkpoint_sha256": "fbd19ff91075877839bb76b022d1b5ad018a980a26603ea9cb4cca7aac0feb6a", + "original_checkpoint_filename": "best_step0198000.pt", + "released_checkpoint_sha256": "9c59bfca391faf96e6d3f130d27dd75f4ed5f0eb1b950ed75361b343e926af14", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "supersedes": { + "step": 184000, + "sha256": "261daa7899501229143136c46a501097caba9ac903650bc9c475c3ee049407b0", + "preserve_historical": true + }, + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K1.json", + "config_sha256": "a2c358f669574dafa25403b5d6e4a0ee0fc3cf237e03399886761f95fd7c0fbd", + "inference_model_id": "Part_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt new file mode 100644 index 0000000000000000000000000000000000000000..454ada470962b92d8ed1efcd38d5090786e0dec1 --- /dev/null +++ b/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c59bfca391faf96e6d3f130d27dd75f4ed5f0eb1b950ed75361b343e926af14 +size 109889741 diff --git a/configs/conditioned/K1.json b/configs/conditioned/K1.json new file mode 100644 index 0000000000000000000000000000000000000000..7e783d66b595ed86dd36a7a92970054414688778 --- /dev/null +++ b/configs/conditioned/K1.json @@ -0,0 +1,31 @@ +{ + "conditioning": "role", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": true, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Bar", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Lead", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/K5.json b/configs/conditioned/K5.json new file mode 100644 index 0000000000000000000000000000000000000000..6bfb9e4f3e77a157c787fb28bc64ac92f1a0d8ea --- /dev/null +++ b/configs/conditioned/K5.json @@ -0,0 +1,31 @@ +{ + "conditioning": "enrollment", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Bar", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Lead", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/K7.json b/configs/conditioned/K7.json new file mode 100644 index 0000000000000000000000000000000000000000..c94d643c4cce80f3cba42ff316f7e10f84afc11d --- /dev/null +++ b/configs/conditioned/K7.json @@ -0,0 +1,31 @@ +{ + "conditioning": "dual", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Bar", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Lead", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/L1.json b/configs/conditioned/L1.json new file mode 100644 index 0000000000000000000000000000000000000000..d71a7b46c450d52faad1d34a7a67c4273db35908 --- /dev/null +++ b/configs/conditioned/L1.json @@ -0,0 +1,31 @@ +{ + "conditioning": "role", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": true, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Lead", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Bar", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/L3.json b/configs/conditioned/L3.json new file mode 100644 index 0000000000000000000000000000000000000000..bf492e7830d5aa3d2507f2e1301ab69702e26048 --- /dev/null +++ b/configs/conditioned/L3.json @@ -0,0 +1,31 @@ +{ + "conditioning": "enrollment", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Lead", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Bar", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/L4.json b/configs/conditioned/L4.json new file mode 100644 index 0000000000000000000000000000000000000000..4d8e836421ad6941765314e4507b7c398fcb7f3c --- /dev/null +++ b/configs/conditioned/L4.json @@ -0,0 +1,31 @@ +{ + "conditioning": "dual", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Lead", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Bar", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/M3.json b/configs/conditioned/M3.json new file mode 100644 index 0000000000000000000000000000000000000000..729d3d97985471978ccbb56a4f0aec32d0ef51f7 --- /dev/null +++ b/configs/conditioned/M3.json @@ -0,0 +1,32 @@ +{ + "conditioning": "enrollment", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Bar", + "Lead", + "S", + "T", + "VP" + ], + "untrained_zero_role": null, + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/Singer_conditioned_htdemucs_SK_full_remix.json b/configs/conditioned/Singer_conditioned_htdemucs_SK_full_remix.json new file mode 100644 index 0000000000000000000000000000000000000000..6bfb9e4f3e77a157c787fb28bc64ac92f1a0d8ea --- /dev/null +++ b/configs/conditioned/Singer_conditioned_htdemucs_SK_full_remix.json @@ -0,0 +1,31 @@ +{ + "conditioning": "enrollment", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Bar", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Lead", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/Singer_conditioned_htdemucs_SK_single_clone.json b/configs/conditioned/Singer_conditioned_htdemucs_SK_single_clone.json new file mode 100644 index 0000000000000000000000000000000000000000..6bfb9e4f3e77a157c787fb28bc64ac92f1a0d8ea --- /dev/null +++ b/configs/conditioned/Singer_conditioned_htdemucs_SK_single_clone.json @@ -0,0 +1,31 @@ +{ + "conditioning": "enrollment", + "source_mode": "target_residual", + "channels": "stereo", + "seconds": 10.0, + "enroll_seconds": 10.0, + "emb_dim": 192, + "role_embedding_dim": 16, + "embedding_backend": "wespeaker_redimnet2", + "freeze_external_embedding": false, + "ht_channels": 48, + "ht_depth": 4, + "ht_growth": 2, + "ht_bottom_channels": 0, + "ht_dconv_mode": 3, + "ht_t_layers": 5, + "ht_nfft": 4096, + "sample_rate": 44100, + "hop_seconds": 4.0, + "training_roles": [ + "A", + "B", + "Bar", + "S", + "T", + "VP" + ], + "untrained_zero_role": "Lead", + "encoder_config": "configs/conditioned/redimnet2.yaml", + "encoder_config_sha256": "57a348c24d56f7bf26794f8bcca2765984726ab56363b2d3e5ef3a27b56524fc" +} diff --git a/configs/conditioned/redimnet2.yaml b/configs/conditioned/redimnet2.yaml new file mode 100644 index 0000000000000000000000000000000000000000..3b2b74882efa1d2df9cb851fd52d6385d315db75 --- /dev/null +++ b/configs/conditioned/redimnet2.yaml @@ -0,0 +1,147 @@ +checkpoint: exp/redimnet2/expB6_v2/-LM/models/model_0.pt +data_type: shard +dataloader_args: + batch_size: 16 + drop_last: true + num_workers: 4 + pin_memory: false + prefetch_factor: 4 +dataset_args: + aug_prob: 0.0 + cmvn: false + filter: false + frontend: tfmel + num_frms: 600 + resample_rate: 16000 + sample_num_per_epoch: 0 + shuffle: true + shuffle_args: + shuffle_size: 2500 + spec_aug: false + speed_perturb: false + tfmel_args: + do_preemph: true + do_spec_aug: false + eps: 1.0e-08 + f_max: 7600 + f_min: 20 + hop_length: 160 + n_fft: 512 + n_mels: 72 + norm_signal: true + win_length: 400 +enable_amp: true +exp_dir: exp/redimnet2/expB6_v2/-LM +gpus: +- 2 +- 3 +grad_accum_steps: 8 +log_batch_interval: 100 +loss: CrossEntropyLoss +loss_args: {} +margin_scheduler: MarginScheduler +margin_update: + epoch_iter: 34125 + final_margin: 0.3 + fix_start_epoch: 0 + increase_start_epoch: 0 + increase_type: exp + initial_margin: 0.3 + update_margin: false +model: ReDimNet2Wrap +model_args: + C: 64 + F: 72 + block_1d_type: conv+att + block_2d_type: basic_resnet + causal: none + compress_tconvs: true + emb_bn: false + embed_dim: 192 + feat_dim: 72 + fm_weigthing_type: NC + global_context_att: true + group_divisor: 1 + hop_length: 160 + out_channels: 224 + pooling_func: ASTP + return_2d_output: true + spec: null + spec_in_channels: 1 + stages_setup: + - - - 1 + - 1 + - 3 + - 3 + - - - 3 + - 3 + - 64 + - - - 2 + - 1 + - 4 + - 2 + - - - 3 + - 3 + - 64 + - - - 1 + - 2 + - 5 + - 2 + - - - 3 + - 3 + - 48 + - - - 2 + - 1 + - 5 + - 1 + - - - 3 + - 3 + - 48 + - - - 1 + - 2 + - 4 + - 0.75 + - - - 3 + - 3 + - 32 + - - - 2 + - 1 + - 3 + - 0.5 + - - - 3 + - 3 + - 24 +model_init: null +noise_data: data/voxceleb/musan/lmdb +num_avg: 1 +num_epochs: 5 +optimizer: SGD +optimizer_args: + lr: 0.0001 + momentum: 0.9 + nesterov: true + weight_decay: 2.0e-05 +projection_args: + do_lm: false + easy_margin: false + embed_dim: 192 + lanbuda: 0.7 + margin_type: C + num_class: 5994 + project_type: sphereface2 + scale: 32.0 + t: 3 +reverb_data: data/voxceleb/rirs/lmdb +save_epoch_interval: 1 +scheduler: ExponentialDecrease +scheduler_args: + epoch_iter: 34125 + final_lr: 1.0e-05 + initial_lr: 0.0001 + num_epochs: 5 + scale_ratio: 4.0 + warm_from_zero: false + warm_up_epoch: 0 +seed: 42 +train_data: data/voxceleb/vox2_dev/shard.list +train_label: data/voxceleb/vox2_dev/utt2spk diff --git a/configs/fixed/JC.yaml b/configs/fixed/JC.yaml new file mode 100644 index 0000000000000000000000000000000000000000..ae15d518eddb38adc4159923be7a4445a824a607 --- /dev/null +++ b/configs/fixed/JC.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['vp', 'bass', 'soprano', 'alto', 'tenor', 'lead_vocal'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/fixed/SK.yaml b/configs/fixed/SK.yaml new file mode 100644 index 0000000000000000000000000000000000000000..ea409e05a43744ef14c7fe3144936622ddcfc7e4 --- /dev/null +++ b/configs/fixed/SK.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['vp', 'bass', 'soprano', 'alto', 'tenor', 'baritone'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/fixed/combined_ep333.yaml b/configs/fixed/combined_ep333.yaml new file mode 100644 index 0000000000000000000000000000000000000000..1221953c7ba30757d69c1e97cf1829b036a52592 --- /dev/null +++ b/configs/fixed/combined_ep333.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['vp', 'bass', 'soprano', 'alto', 'tenor', 'other'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/JC/alto.yaml b/configs/specialists/JC/alto.yaml new file mode 100644 index 0000000000000000000000000000000000000000..5dc383b6614ec5455d9e9c760e4798a141b39605 --- /dev/null +++ b/configs/specialists/JC/alto.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonalto', 'alto'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/JC/bass.yaml b/configs/specialists/JC/bass.yaml new file mode 100644 index 0000000000000000000000000000000000000000..2928ae11628bf7909e40564851cbda626f553a44 --- /dev/null +++ b/configs/specialists/JC/bass.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonbass', 'bass'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/JC/lead_vocal.yaml b/configs/specialists/JC/lead_vocal.yaml new file mode 100644 index 0000000000000000000000000000000000000000..71e4f19d677f1754777489613d9d5f56d8d70036 --- /dev/null +++ b/configs/specialists/JC/lead_vocal.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonlead_vocal', 'lead_vocal'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/JC/soprano.yaml b/configs/specialists/JC/soprano.yaml new file mode 100644 index 0000000000000000000000000000000000000000..e6edab5b9be16a5e153217050ba5435744d08ccd --- /dev/null +++ b/configs/specialists/JC/soprano.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonsoprano', 'soprano'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/JC/tenor.yaml b/configs/specialists/JC/tenor.yaml new file mode 100644 index 0000000000000000000000000000000000000000..3ad13560b17ac48b4cfd75f601491a3e4b570b97 --- /dev/null +++ b/configs/specialists/JC/tenor.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nontenor', 'tenor'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/JC/vp.yaml b/configs/specialists/JC/vp.yaml new file mode 100644 index 0000000000000000000000000000000000000000..dab7b6a33443c814c615fba2b83f442c1267c665 --- /dev/null +++ b/configs/specialists/JC/vp.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonvp', 'vp'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/SK/alto.yaml b/configs/specialists/SK/alto.yaml new file mode 100644 index 0000000000000000000000000000000000000000..5dc383b6614ec5455d9e9c760e4798a141b39605 --- /dev/null +++ b/configs/specialists/SK/alto.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonalto', 'alto'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/SK/baritone.yaml b/configs/specialists/SK/baritone.yaml new file mode 100644 index 0000000000000000000000000000000000000000..64b6b726f47e4fef29555ebe9cdd398fb83a06f2 --- /dev/null +++ b/configs/specialists/SK/baritone.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonbaritone', 'baritone'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/SK/bass.yaml b/configs/specialists/SK/bass.yaml new file mode 100644 index 0000000000000000000000000000000000000000..2928ae11628bf7909e40564851cbda626f553a44 --- /dev/null +++ b/configs/specialists/SK/bass.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonbass', 'bass'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/SK/soprano.yaml b/configs/specialists/SK/soprano.yaml new file mode 100644 index 0000000000000000000000000000000000000000..e6edab5b9be16a5e153217050ba5435744d08ccd --- /dev/null +++ b/configs/specialists/SK/soprano.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonsoprano', 'soprano'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/SK/tenor.yaml b/configs/specialists/SK/tenor.yaml new file mode 100644 index 0000000000000000000000000000000000000000..3ad13560b17ac48b4cfd75f601491a3e4b570b97 --- /dev/null +++ b/configs/specialists/SK/tenor.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nontenor', 'tenor'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/configs/specialists/SK/vp.yaml b/configs/specialists/SK/vp.yaml new file mode 100644 index 0000000000000000000000000000000000000000..dab7b6a33443c814c615fba2b83f442c1267c665 --- /dev/null +++ b/configs/specialists/SK/vp.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['nonvp', 'vp'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt b/fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt new file mode 100644 index 0000000000000000000000000000000000000000..663844f016f8a6380f9d73d3e9d505aa7985d03c --- /dev/null +++ b/fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:432c3806e03474a8fe15e586b1083e977c22751a3099ff82625c16a26490907d +size 109830677 diff --git a/fixed/JC/legacy/model.json b/fixed/JC/legacy/model.json new file mode 100644 index 0000000000000000000000000000000000000000..df8672fc89c24eccc2b07fd511205c3905c5afbd --- /dev/null +++ b/fixed/JC/legacy/model.json @@ -0,0 +1,42 @@ +{ + "id": "Fixed_6stem_htdemucs_JC", + "family": "fixed", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": null, + "epoch": 321, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt", + "original_checkpoint_filename": "model_htdemucs_ep_321_sdr_7.9284_si_sdr_6.9219_k_sdr_94.9111.ckpt", + "original_checkpoint_sha256": "1d65d49aaa7fc52745e4c476d14d90d7bddb942765c169e92bf045ea33d4ab6b", + "released_checkpoint_sha256": "432c3806e03474a8fe15e586b1083e977c22751a3099ff82625c16a26490907d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt", + "weight_license": "cc-by-nc-4.0", + "source_order": [ + "vp", + "bass", + "soprano", + "alto", + "tenor", + "lead_vocal" + ], + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "display_name": "Fixed_6stem_htdemucs_JC", + "aliases": [ + "fixed-jc-ep321" + ], + "previous_checkpoint_path": "fixed/JC/legacy/epoch_0321.pt", + "public_inference_status": "published_export_strict_loading_verified", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/fixed/JC.yaml", + "config_sha256": "b3e5ff56161510e5abd8236eabb0b020446b990bc013a311ef10e640c01bad1e", + "inference_model_id": "Fixed_6stem_htdemucs_JC", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109830677, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/fixed/README.md b/fixed/README.md new file mode 100644 index 0000000000000000000000000000000000000000..c830977d86cff71978fa909ccde0226fb644c8f7 --- /dev/null +++ b/fixed/README.md @@ -0,0 +1,5 @@ +# Fixed-output models + +Fixed_6stem_htdemucs_SKnJC is available as a validated model-only [checkpoint](SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt). +**Weight license: CC BY-NC 4.0**; see [weight notices](../WEIGHTS_NOTICE.md). +Fixed_6stem_htdemucs_JC and Fixed_6stem_htdemucs_SK are also available as tensor-only inference exports. Paths and SHA-256 hashes are in [manifest.json](../manifest.json). diff --git a/fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt b/fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt new file mode 100644 index 0000000000000000000000000000000000000000..28675eef6995df0752f076ab68b3bf8c9a91be4b --- /dev/null +++ b/fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e82eced15d0466a3ada6584e28188975841930d21153749fe163ad07af19d38c +size 109830677 diff --git a/fixed/SK/legacy/model.json b/fixed/SK/legacy/model.json new file mode 100644 index 0000000000000000000000000000000000000000..ee8f172627328fc9d7d15786cbb36286318ab3cd --- /dev/null +++ b/fixed/SK/legacy/model.json @@ -0,0 +1,42 @@ +{ + "id": "Fixed_6stem_htdemucs_SK", + "family": "fixed", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": null, + "epoch": 279, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt", + "original_checkpoint_filename": "model_htdemucs_ep_279_sdr_-7.7538_si_sdr_-5.3812_k_sdr_81.6959.ckpt", + "original_checkpoint_sha256": "51c1a71b22d82c34e31f41e314fc9a84055d9f55f5d77c8c36b4280a59a19f0b", + "released_checkpoint_sha256": "e82eced15d0466a3ada6584e28188975841930d21153749fe163ad07af19d38c", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt", + "weight_license": "cc-by-nc-4.0", + "source_order": [ + "vp", + "bass", + "soprano", + "alto", + "tenor", + "baritone" + ], + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "display_name": "Fixed_6stem_htdemucs_SK", + "aliases": [ + "fixed-sk-ep279" + ], + "previous_checkpoint_path": "fixed/SK/legacy/epoch_0279.pt", + "public_inference_status": "published_export_strict_loading_verified", + "config": "configs/fixed/SK.yaml", + "config_sha256": "2210240362436d809e4f9a037cf1e40146206b906b8e95b9f28144d1da84e81d", + "inference_model_id": "Fixed_6stem_htdemucs_SK", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109830677, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt b/fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt new file mode 100644 index 0000000000000000000000000000000000000000..c0c8bc2863370ac73894f6508052b4958cf36246 --- /dev/null +++ b/fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f87f3a1224de680085fea0c186600edc7cfbb91b8f56f8862da4772565e0cda +size 109822110 diff --git a/fixed/SKnJC/legacy/NOTICE.md b/fixed/SKnJC/legacy/NOTICE.md new file mode 100644 index 0000000000000000000000000000000000000000..0c0497704ce5fdbe5c3bd0e622cb5e6c12164009 --- /dev/null +++ b/fixed/SKnJC/legacy/NOTICE.md @@ -0,0 +1,17 @@ +# Fixed_6stem_htdemucs_SKnJC weight status + +**Project weight license: CC BY-NC 4.0.** See the [weight notices](../../../WEIGHTS_NOTICE.md) for scope and upstream rights. + +The fine-tuned model modifies the initialization through training on JC and SK; +the tensor-only export removes training/optimizer state. The checkpoint is unchanged. + +Architecture source code: [Meta Demucs](https://github.com/facebookresearch/demucs), +MIT; retain the [Meta code notice](licenses/Demucs-MIT.txt). Pretrained weight +permissions are separate; see [the author's clarification](https://github.com/facebookresearch/demucs/issues/327#issuecomment-1134828611). +Training and inference code: [Music-Source-Separation-Training](https://github.com/ZFTurbo/Music-Source-Separation-Training), +MIT; retain the [MSS notice](licenses/MSS-MIT.txt). + +The research PT6S initializer matches official htdemucs_6s signature 5c90dfd2: +all 525 tensors match after float16-to-float32 conversion. +[Provenance](https://github.com/pting126/ACappellaSet/blob/main/third_party/initialization.json). +No dataset audio or other model weights are licensed by this notice. diff --git a/fixed/SKnJC/legacy/config.yaml b/fixed/SKnJC/legacy/config.yaml new file mode 100644 index 0000000000000000000000000000000000000000..1221953c7ba30757d69c1e97cf1829b036a52592 --- /dev/null +++ b/fixed/SKnJC/legacy/config.yaml @@ -0,0 +1,104 @@ +audio: + chunk_size: 441000 # 44100 * 10s + min_mean_abs: 0.001 + hop_length: 1024 + +training: + batch_size: 8 + gradient_accumulation_steps: 1 + grad_clip: 0 + segment: 10 + shift: 1 + samplerate: 44100 + channels: 2 + normalize: true + instruments: ['vp', 'bass', 'soprano', 'alto', 'tenor', 'other'] + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + optimizer: adam + lr: 9.0e-05 + patience: 2 + reduce_factor: 0.95 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + other_fix: false + use_amp: true + +augmentations: + enable: true + loudness: true + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true + mixup_probs: [0.2, 0.02] + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + all: + channel_shuffle: 0.5 + random_inverse: 0.1 + random_polarity: 0.5 + +inference: + num_overlap: 4 + batch_size: 8 + +model: htdemucs + +htdemucs: + channels: 48 + channels_time: + growth: 2 + num_subbands: 1 + nfft: 4096 + wiener_iters: 0 + end_iters: 0 + wiener_residual: false + cac: true + depth: 4 + rewrite: true + multi_freqs: [] + multi_freqs_depth: 3 + freq_emb: 0.2 + emb_scale: 10 + emb_smooth: true + kernel_size: 8 + stride: 4 + time_stride: 2 + context: 1 + context_enc: 0 + norm_starts: 4 + norm_groups: 4 + dconv_mode: 3 + dconv_depth: 2 + dconv_comp: 8 + dconv_init: 1e-3 + bottom_channels: 0 + t_layers: 5 + t_hidden_scale: 4.0 + t_heads: 8 + t_dropout: 0.0 + t_layer_scale: True + t_gelu: True + t_emb: sin + t_max_positions: 10000 + t_max_period: 10000.0 + t_weight_pos_embed: 1.0 + t_cape_mean_normalize: True + t_cape_augment: True + t_cape_glob_loc_scale: [5000.0, 1.0, 1.4] + t_sin_random_shift: 0 + t_norm_in: True + t_norm_in_group: False + t_group_norm: False + t_norm_first: True + t_norm_out: True + t_weight_decay: 0.0 + t_lr: + t_sparse_self_attn: False + t_sparse_cross_attn: False + t_mask_type: diag + t_mask_random_seed: 42 + t_sparse_attn_window: 400 + t_global_window: 100 diff --git a/fixed/SKnJC/legacy/licenses/Demucs-MIT.txt b/fixed/SKnJC/legacy/licenses/Demucs-MIT.txt new file mode 100644 index 0000000000000000000000000000000000000000..b93be90515ccd0b9daedaa589e42bf5929693f1f --- /dev/null +++ b/fixed/SKnJC/legacy/licenses/Demucs-MIT.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) Meta Platforms, Inc. and affiliates. + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/fixed/SKnJC/legacy/licenses/MSS-MIT.txt b/fixed/SKnJC/legacy/licenses/MSS-MIT.txt new file mode 100644 index 0000000000000000000000000000000000000000..9d7186e88bca9975edd65956cd499fa60bd04251 --- /dev/null +++ b/fixed/SKnJC/legacy/licenses/MSS-MIT.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2024 Roman Solovyev (ZFTurbo) + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/fixed/SKnJC/legacy/model.json b/fixed/SKnJC/legacy/model.json new file mode 100644 index 0000000000000000000000000000000000000000..f4845f6918b92ed4a0fb1eb5c7ee07f07312800e --- /dev/null +++ b/fixed/SKnJC/legacy/model.json @@ -0,0 +1,54 @@ +{ + "id": "Fixed_6stem_htdemucs_SKnJC", + "family": "fixed", + "training_corpus": "Combined", + "recipe": "legacy", + "target_role": null, + "epoch": 333, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt", + "original_checkpoint_filename": "model_htdemucs_ep_333_sdr_-1.2322_si_sdr_-0.3075_k_sdr_87.1730.ckpt", + "original_checkpoint_sha256": "cbfa201798f7e7f4766bacd3aa8622b3ae40321729bd7f0b3555ac4348c57879", + "released_checkpoint_sha256": "8f87f3a1224de680085fea0c186600edc7cfbb91b8f56f8862da4772565e0cda", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt", + "weight_license": "cc-by-nc-4.0", + "source_order": [ + "vp", + "bass", + "soprano", + "alto", + "tenor", + "other" + ], + "code_support": "Original and portable Combined checkpoints validated on Linux CPU; other platforms pending.", + "sixth_output_note": "other represents JC Lead / SK Baritone training semantics; not two independent outputs.", + "config_path": "fixed/SKnJC/legacy/config.yaml", + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "prepared_export": { + "export_sha256": "8f87f3a1224de680085fea0c186600edc7cfbb91b8f56f8862da4772565e0cda", + "export_bytes": 109822110, + "format": "pytorch_state_dict", + "weights_only": true, + "tensor_count": 525, + "tensor_values_bit_identical": true, + "optimizer_included": false, + "validation": "Linux CPU: two clips, all six outputs bit-identical to original checkpoint and research reference; six behavioral tests passed" + }, + "released_checkpoint_bytes": 109822110, + "notice_path": "WEIGHTS_NOTICE.md", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "display_name": "Fixed_6stem_htdemucs_SKnJC", + "aliases": [ + "fixed-combined-ep333" + ], + "previous_checkpoint_path": "fixed/Combined/legacy/epoch_0333.pt", + "public_inference_status": "published_export_strict_loading_verified", + "config": "configs/fixed/combined_ep333.yaml", + "config_sha256": "55bd3b4080fb20bfe445647bef934c991e475b92b64401c339f9610f34eb2244", + "inference_model_id": "Fixed_6stem_htdemucs_SKnJC", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/licenses/CC-BY-NC-4.0.txt b/licenses/CC-BY-NC-4.0.txt new file mode 100644 index 0000000000000000000000000000000000000000..fe463e0f7888bbf8b82e42d55f5743508ddafb7e --- /dev/null +++ b/licenses/CC-BY-NC-4.0.txt @@ -0,0 +1,408 @@ +Attribution-NonCommercial 4.0 International + +======================================================================= + +Creative Commons Corporation ("Creative Commons") is not a law firm and +does not provide legal services or legal advice. Distribution of +Creative Commons public licenses does not create a lawyer-client or +other relationship. Creative Commons makes its licenses and related +information available on an "as-is" basis. Creative Commons gives no +warranties regarding its licenses, any material licensed under their +terms and conditions, or any related information. Creative Commons +disclaims all liability for damages resulting from their use to the +fullest extent possible. + +Using Creative Commons Public Licenses + +Creative Commons public licenses provide a standard set of terms and +conditions that creators and other rights holders may use to share +original works of authorship and other material subject to copyright +and certain other rights specified in the public license below. The +following considerations are for informational purposes only, are not +exhaustive, and do not form part of our licenses. + + Considerations for licensors: Our public licenses are + intended for use by those authorized to give the public + permission to use material in ways otherwise restricted by + copyright and certain other rights. Our licenses are + irrevocable. Licensors should read and understand the terms + and conditions of the license they choose before applying it. + Licensors should also secure all rights necessary before + applying our licenses so that the public can reuse the + material as expected. Licensors should clearly mark any + material not subject to the license. This includes other CC- + licensed material, or material used under an exception or + limitation to copyright. More considerations for licensors: + wiki.creativecommons.org/Considerations_for_licensors + + Considerations for the public: By using one of our public + licenses, a licensor grants the public permission to use the + licensed material under specified terms and conditions. If + the licensor's permission is not necessary for any reason--for + example, because of any applicable exception or limitation to + copyright--then that use is not regulated by the license. Our + licenses grant only permissions under copyright and certain + other rights that a licensor has authority to grant. Use of + the licensed material may still be restricted for other + reasons, including because others have copyright or other + rights in the material. A licensor may make special requests, + such as asking that all changes be marked or described. + Although not required by our licenses, you are encouraged to + respect those requests where reasonable. More considerations + for the public: + wiki.creativecommons.org/Considerations_for_licensees + +======================================================================= + +Creative Commons Attribution-NonCommercial 4.0 International Public +License + +By exercising the Licensed Rights (defined below), You accept and agree +to be bound by the terms and conditions of this Creative Commons +Attribution-NonCommercial 4.0 International Public License ("Public +License"). To the extent this Public License may be interpreted as a +contract, You are granted the Licensed Rights in consideration of Your +acceptance of these terms and conditions, and the Licensor grants You +such rights in consideration of benefits the Licensor receives from +making the Licensed Material available under these terms and +conditions. + + +Section 1 -- Definitions. + + a. Adapted Material means material subject to Copyright and Similar + Rights that is derived from or based upon the Licensed Material + and in which the Licensed Material is translated, altered, + arranged, transformed, or otherwise modified in a manner requiring + permission under the Copyright and Similar Rights held by the + Licensor. For purposes of this Public License, where the Licensed + Material is a musical work, performance, or sound recording, + Adapted Material is always produced where the Licensed Material is + synched in timed relation with a moving image. + + b. Adapter's License means the license You apply to Your Copyright + and Similar Rights in Your contributions to Adapted Material in + accordance with the terms and conditions of this Public License. + + c. Copyright and Similar Rights means copyright and/or similar rights + closely related to copyright including, without limitation, + performance, broadcast, sound recording, and Sui Generis Database + Rights, without regard to how the rights are labeled or + categorized. For purposes of this Public License, the rights + specified in Section 2(b)(1)-(2) are not Copyright and Similar + Rights. + d. Effective Technological Measures means those measures that, in the + absence of proper authority, may not be circumvented under laws + fulfilling obligations under Article 11 of the WIPO Copyright + Treaty adopted on December 20, 1996, and/or similar international + agreements. + + e. Exceptions and Limitations means fair use, fair dealing, and/or + any other exception or limitation to Copyright and Similar Rights + that applies to Your use of the Licensed Material. + + f. Licensed Material means the artistic or literary work, database, + or other material to which the Licensor applied this Public + License. + + g. Licensed Rights means the rights granted to You subject to the + terms and conditions of this Public License, which are limited to + all Copyright and Similar Rights that apply to Your use of the + Licensed Material and that the Licensor has authority to license. + + h. Licensor means the individual(s) or entity(ies) granting rights + under this Public License. + + i. NonCommercial means not primarily intended for or directed towards + commercial advantage or monetary compensation. For purposes of + this Public License, the exchange of the Licensed Material for + other material subject to Copyright and Similar Rights by digital + file-sharing or similar means is NonCommercial provided there is + no payment of monetary compensation in connection with the + exchange. + + j. Share means to provide material to the public by any means or + process that requires permission under the Licensed Rights, such + as reproduction, public display, public performance, distribution, + dissemination, communication, or importation, and to make material + available to the public including in ways that members of the + public may access the material from a place and at a time + individually chosen by them. + + k. Sui Generis Database Rights means rights other than copyright + resulting from Directive 96/9/EC of the European Parliament and of + the Council of 11 March 1996 on the legal protection of databases, + as amended and/or succeeded, as well as other essentially + equivalent rights anywhere in the world. + + l. You means the individual or entity exercising the Licensed Rights + under this Public License. Your has a corresponding meaning. + + +Section 2 -- Scope. + + a. License grant. + + 1. Subject to the terms and conditions of this Public License, + the Licensor hereby grants You a worldwide, royalty-free, + non-sublicensable, non-exclusive, irrevocable license to + exercise the Licensed Rights in the Licensed Material to: + + a. reproduce and Share the Licensed Material, in whole or + in part, for NonCommercial purposes only; and + + b. produce, reproduce, and Share Adapted Material for + NonCommercial purposes only. + + 2. Exceptions and Limitations. For the avoidance of doubt, where + Exceptions and Limitations apply to Your use, this Public + License does not apply, and You do not need to comply with + its terms and conditions. + + 3. Term. The term of this Public License is specified in Section + 6(a). + + 4. Media and formats; technical modifications allowed. The + Licensor authorizes You to exercise the Licensed Rights in + all media and formats whether now known or hereafter created, + and to make technical modifications necessary to do so. The + Licensor waives and/or agrees not to assert any right or + authority to forbid You from making technical modifications + necessary to exercise the Licensed Rights, including + technical modifications necessary to circumvent Effective + Technological Measures. For purposes of this Public License, + simply making modifications authorized by this Section 2(a) + (4) never produces Adapted Material. + + 5. Downstream recipients. + + a. Offer from the Licensor -- Licensed Material. Every + recipient of the Licensed Material automatically + receives an offer from the Licensor to exercise the + Licensed Rights under the terms and conditions of this + Public License. + + b. No downstream restrictions. You may not offer or impose + any additional or different terms or conditions on, or + apply any Effective Technological Measures to, the + Licensed Material if doing so restricts exercise of the + Licensed Rights by any recipient of the Licensed + Material. + + 6. No endorsement. Nothing in this Public License constitutes or + may be construed as permission to assert or imply that You + are, or that Your use of the Licensed Material is, connected + with, or sponsored, endorsed, or granted official status by, + the Licensor or others designated to receive attribution as + provided in Section 3(a)(1)(A)(i). + + b. Other rights. + + 1. Moral rights, such as the right of integrity, are not + licensed under this Public License, nor are publicity, + privacy, and/or other similar personality rights; however, to + the extent possible, the Licensor waives and/or agrees not to + assert any such rights held by the Licensor to the limited + extent necessary to allow You to exercise the Licensed + Rights, but not otherwise. + + 2. Patent and trademark rights are not licensed under this + Public License. + + 3. To the extent possible, the Licensor waives any right to + collect royalties from You for the exercise of the Licensed + Rights, whether directly or through a collecting society + under any voluntary or waivable statutory or compulsory + licensing scheme. In all other cases the Licensor expressly + reserves any right to collect such royalties, including when + the Licensed Material is used other than for NonCommercial + purposes. + + +Section 3 -- License Conditions. + +Your exercise of the Licensed Rights is expressly made subject to the +following conditions. + + a. Attribution. + + 1. If You Share the Licensed Material (including in modified + form), You must: + + a. retain the following if it is supplied by the Licensor + with the Licensed Material: + + i. identification of the creator(s) of the Licensed + Material and any others designated to receive + attribution, in any reasonable manner requested by + the Licensor (including by pseudonym if + designated); + + ii. a copyright notice; + + iii. a notice that refers to this Public License; + + iv. a notice that refers to the disclaimer of + warranties; + + v. a URI or hyperlink to the Licensed Material to the + extent reasonably practicable; + + b. indicate if You modified the Licensed Material and + retain an indication of any previous modifications; and + + c. indicate the Licensed Material is licensed under this + Public License, and include the text of, or the URI or + hyperlink to, this Public License. + + 2. You may satisfy the conditions in Section 3(a)(1) in any + reasonable manner based on the medium, means, and context in + which You Share the Licensed Material. For example, it may be + reasonable to satisfy the conditions by providing a URI or + hyperlink to a resource that includes the required + information. + + 3. If requested by the Licensor, You must remove any of the + information required by Section 3(a)(1)(A) to the extent + reasonably practicable. + + 4. If You Share Adapted Material You produce, the Adapter's + License You apply must not prevent recipients of the Adapted + Material from complying with this Public License. + + +Section 4 -- Sui Generis Database Rights. + +Where the Licensed Rights include Sui Generis Database Rights that +apply to Your use of the Licensed Material: + + a. for the avoidance of doubt, Section 2(a)(1) grants You the right + to extract, reuse, reproduce, and Share all or a substantial + portion of the contents of the database for NonCommercial purposes + only; + + b. if You include all or a substantial portion of the database + contents in a database in which You have Sui Generis Database + Rights, then the database in which You have Sui Generis Database + Rights (but not its individual contents) is Adapted Material; and + + c. You must comply with the conditions in Section 3(a) if You Share + all or a substantial portion of the contents of the database. + +For the avoidance of doubt, this Section 4 supplements and does not +replace Your obligations under this Public License where the Licensed +Rights include other Copyright and Similar Rights. + + +Section 5 -- Disclaimer of Warranties and Limitation of Liability. + + a. UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE + EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS + AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF + ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS, + IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION, + WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR + PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS, + ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT + KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT + ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU. + + b. TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE + TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION, + NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT, + INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES, + COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR + USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN + ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR + DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR + IN PART, THIS LIMITATION MAY NOT APPLY TO YOU. + + c. The disclaimer of warranties and limitation of liability provided + above shall be interpreted in a manner that, to the extent + possible, most closely approximates an absolute disclaimer and + waiver of all liability. + + +Section 6 -- Term and Termination. + + a. This Public License applies for the term of the Copyright and + Similar Rights licensed here. However, if You fail to comply with + this Public License, then Your rights under this Public License + terminate automatically. + + b. Where Your right to use the Licensed Material has terminated under + Section 6(a), it reinstates: + + 1. automatically as of the date the violation is cured, provided + it is cured within 30 days of Your discovery of the + violation; or + + 2. upon express reinstatement by the Licensor. + + For the avoidance of doubt, this Section 6(b) does not affect any + right the Licensor may have to seek remedies for Your violations + of this Public License. + + c. For the avoidance of doubt, the Licensor may also offer the + Licensed Material under separate terms or conditions or stop + distributing the Licensed Material at any time; however, doing so + will not terminate this Public License. + + d. Sections 1, 5, 6, 7, and 8 survive termination of this Public + License. + + +Section 7 -- Other Terms and Conditions. + + a. The Licensor shall not be bound by any additional or different + terms or conditions communicated by You unless expressly agreed. + + b. Any arrangements, understandings, or agreements regarding the + Licensed Material not stated herein are separate from and + independent of the terms and conditions of this Public License. + + +Section 8 -- Interpretation. + + a. For the avoidance of doubt, this Public License does not, and + shall not be interpreted to, reduce, limit, restrict, or impose + conditions on any use of the Licensed Material that could lawfully + be made without permission under this Public License. + + b. To the extent possible, if any provision of this Public License is + deemed unenforceable, it shall be automatically reformed to the + minimum extent necessary to make it enforceable. If the provision + cannot be reformed, it shall be severed from this Public License + without affecting the enforceability of the remaining terms and + conditions. + + c. No term or condition of this Public License will be waived and no + failure to comply consented to unless expressly agreed to by the + Licensor. + + d. Nothing in this Public License constitutes or may be interpreted + as a limitation upon, or waiver of, any privileges and immunities + that apply to the Licensor or You, including from the legal + processes of any jurisdiction or authority. + +======================================================================= + +Creative Commons is not a party to its public +licenses. Notwithstanding, Creative Commons may elect to apply one of +its public licenses to material it publishes and in those instances +will be considered the “Licensor.” The text of the Creative Commons +public licenses is dedicated to the public domain under the CC0 Public +Domain Dedication. Except for the limited purpose of indicating that +material is shared under a Creative Commons public license or as +otherwise permitted by the Creative Commons policies published at +creativecommons.org/policies, Creative Commons does not authorize the +use of the trademark "Creative Commons" or any other trademark or logo +of Creative Commons without its prior written consent including, +without limitation, in connection with any unauthorized modifications +to any of its public licenses or any other arrangements, +understandings, or agreements concerning use of licensed material. For +the avoidance of doubt, this paragraph does not form part of the +public licenses. + +Creative Commons may be contacted at creativecommons.org. + diff --git a/licenses/Demucs-MIT.txt b/licenses/Demucs-MIT.txt new file mode 100644 index 0000000000000000000000000000000000000000..b93be90515ccd0b9daedaa589e42bf5929693f1f --- /dev/null +++ b/licenses/Demucs-MIT.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) Meta Platforms, Inc. and affiliates. + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/licenses/MSS-MIT.txt b/licenses/MSS-MIT.txt new file mode 100644 index 0000000000000000000000000000000000000000..9d7186e88bca9975edd65956cd499fa60bd04251 --- /dev/null +++ b/licenses/MSS-MIT.txt @@ -0,0 +1,21 @@ +MIT License + +Copyright (c) 2024 Roman Solovyev (ZFTurbo) + +Permission is hereby granted, free of charge, to any person obtaining a copy +of this software and associated documentation files (the "Software"), to deal +in the Software without restriction, including without limitation the rights +to use, copy, modify, merge, publish, distribute, sublicense, and/or sell +copies of the Software, and to permit persons to whom the Software is +furnished to do so, subject to the following conditions: + +The above copyright notice and this permission notice shall be included in all +copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, +OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE +SOFTWARE. diff --git a/licenses/WeSpeaker-Apache-2.0.txt b/licenses/WeSpeaker-Apache-2.0.txt new file mode 100644 index 0000000000000000000000000000000000000000..d645695673349e3947e8e5ae42332d0ac3164cd7 --- /dev/null +++ b/licenses/WeSpeaker-Apache-2.0.txt @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/manifest.json b/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..973b708c2a969eba11718cadaf731f2e0ea622ec --- /dev/null +++ b/manifest.json @@ -0,0 +1,1124 @@ +{ + "schema_version": 1, + "status": "private_all_inference_exports_validated", + "code_repository": "https://github.com/pting126/ACappellaSet", + "dataset_hosting": "https://huggingface.co/datasets/AcappellaSet/ACappellaSet (released; public Dataset Card with manually approved Non-commercial Research Use access)", + "models": [ + { + "id": "Fixed_6stem_htdemucs_JC", + "family": "fixed", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": null, + "epoch": 321, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt", + "original_checkpoint_filename": "model_htdemucs_ep_321_sdr_7.9284_si_sdr_6.9219_k_sdr_94.9111.ckpt", + "original_checkpoint_sha256": "1d65d49aaa7fc52745e4c476d14d90d7bddb942765c169e92bf045ea33d4ab6b", + "released_checkpoint_sha256": "432c3806e03474a8fe15e586b1083e977c22751a3099ff82625c16a26490907d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/fixed/JC/legacy/Fixed_6stem_htdemucs_JC.pt", + "weight_license": "cc-by-nc-4.0", + "source_order": [ + "vp", + "bass", + "soprano", + "alto", + "tenor", + "lead_vocal" + ], + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "display_name": "Fixed_6stem_htdemucs_JC", + "aliases": [ + "fixed-jc-ep321" + ], + "previous_checkpoint_path": "fixed/JC/legacy/epoch_0321.pt", + "public_inference_status": "published_export_strict_loading_verified", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/fixed/JC.yaml", + "config_sha256": "b3e5ff56161510e5abd8236eabb0b020446b990bc013a311ef10e640c01bad1e", + "inference_model_id": "Fixed_6stem_htdemucs_JC", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109830677, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "Fixed_6stem_htdemucs_SK", + "family": "fixed", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": null, + "epoch": 279, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt", + "original_checkpoint_filename": "model_htdemucs_ep_279_sdr_-7.7538_si_sdr_-5.3812_k_sdr_81.6959.ckpt", + "original_checkpoint_sha256": "51c1a71b22d82c34e31f41e314fc9a84055d9f55f5d77c8c36b4280a59a19f0b", + "released_checkpoint_sha256": "e82eced15d0466a3ada6584e28188975841930d21153749fe163ad07af19d38c", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/fixed/SK/legacy/Fixed_6stem_htdemucs_SK.pt", + "weight_license": "cc-by-nc-4.0", + "source_order": [ + "vp", + "bass", + "soprano", + "alto", + "tenor", + "baritone" + ], + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "display_name": "Fixed_6stem_htdemucs_SK", + "aliases": [ + "fixed-sk-ep279" + ], + "previous_checkpoint_path": "fixed/SK/legacy/epoch_0279.pt", + "public_inference_status": "published_export_strict_loading_verified", + "config": "configs/fixed/SK.yaml", + "config_sha256": "2210240362436d809e4f9a037cf1e40146206b906b8e95b9f28144d1da84e81d", + "inference_model_id": "Fixed_6stem_htdemucs_SK", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109830677, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "Fixed_6stem_htdemucs_SKnJC", + "family": "fixed", + "training_corpus": "Combined", + "recipe": "legacy", + "target_role": null, + "epoch": 333, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt", + "original_checkpoint_filename": "model_htdemucs_ep_333_sdr_-1.2322_si_sdr_-0.3075_k_sdr_87.1730.ckpt", + "original_checkpoint_sha256": "cbfa201798f7e7f4766bacd3aa8622b3ae40321729bd7f0b3555ac4348c57879", + "released_checkpoint_sha256": "8f87f3a1224de680085fea0c186600edc7cfbb91b8f56f8862da4772565e0cda", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/fixed/SKnJC/legacy/Fixed_6stem_htdemucs_SKnJC.pt", + "weight_license": "cc-by-nc-4.0", + "source_order": [ + "vp", + "bass", + "soprano", + "alto", + "tenor", + "other" + ], + "code_support": "Original and portable Combined checkpoints validated on Linux CPU; other platforms pending.", + "sixth_output_note": "other represents JC Lead / SK Baritone training semantics; not two independent outputs.", + "config_path": "fixed/SKnJC/legacy/config.yaml", + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "prepared_export": { + "export_sha256": "8f87f3a1224de680085fea0c186600edc7cfbb91b8f56f8862da4772565e0cda", + "export_bytes": 109822110, + "format": "pytorch_state_dict", + "weights_only": true, + "tensor_count": 525, + "tensor_values_bit_identical": true, + "optimizer_included": false, + "validation": "Linux CPU: two clips, all six outputs bit-identical to original checkpoint and research reference; six behavioral tests passed" + }, + "released_checkpoint_bytes": 109822110, + "notice_path": "WEIGHTS_NOTICE.md", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "display_name": "Fixed_6stem_htdemucs_SKnJC", + "aliases": [ + "fixed-combined-ep333" + ], + "previous_checkpoint_path": "fixed/Combined/legacy/epoch_0333.pt", + "public_inference_status": "published_export_strict_loading_verified", + "config": "configs/fixed/combined_ep333.yaml", + "config_sha256": "55bd3b4080fb20bfe445647bef934c991e475b92b64401c339f9610f34eb2244", + "inference_model_id": "Fixed_6stem_htdemucs_SKnJC", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-jc-alto-ep223", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "alto", + "epoch": 223, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/alto/epoch_0223.pt", + "original_checkpoint_filename": "model_htdemucs_ep_223_si_sdr_6.0926.ckpt", + "original_checkpoint_sha256": "5c8118424cbe5264b953f37d8336194b66089bac0aa18c1a32d1c41b386e7e1e", + "released_checkpoint_sha256": "9212124f4a3eab2b9a4a14669bc3e52c66de3b1b795e805a5b575e430f24072d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/alto/epoch_0223.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/alto.yaml", + "config_sha256": "a0206c56e8078c7c3443eee4ba343f35baae36975ec403052bb99d0345c1e8b4", + "inference_model_id": "specialists-jc-alto-ep223", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-jc-bass-ep110", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "bass", + "epoch": 110, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/bass/epoch_0110.pt", + "original_checkpoint_filename": "model_htdemucs_ep_110_si_sdr_16.6663.ckpt", + "original_checkpoint_sha256": "23d152aa2ae0250fe852ac5e31ae68f2d70736acd4cb2a5f9515a9b10302a810", + "released_checkpoint_sha256": "7b4b70dd70fa921cfe7d5b79c8e31004c5ea9893a9de659b2f5592fcde4a213c", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/bass/epoch_0110.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/bass.yaml", + "config_sha256": "4f0a1834931f5e2309e28b1d74059b1a77fff3d018037ddfdab7dcd09328c698", + "inference_model_id": "specialists-jc-bass-ep110", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-jc-lead-ep149", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "lead", + "epoch": 149, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/lead/epoch_0149.pt", + "original_checkpoint_filename": "model_htdemucs_ep_149_si_sdr_9.6743.ckpt", + "original_checkpoint_sha256": "0a37e456852527fd2d0c7d983ba086f1ebb15064a39633f99c06cbbb37914c28", + "released_checkpoint_sha256": "4825c5864e9297d2e1bc7977ea5879af522f182b56483581bf14b90b8c18f465", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/lead/epoch_0149.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/lead_vocal.yaml", + "config_sha256": "dcb7bb687bad3a7f00d03491b3fba10fa12a2b5777025cc04c2a68fcfda311e0", + "inference_model_id": "specialists-jc-lead-ep149", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-jc-soprano-ep162", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "soprano", + "epoch": 162, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/soprano/epoch_0162.pt", + "original_checkpoint_filename": "model_htdemucs_ep_162_si_sdr_6.1072.ckpt", + "original_checkpoint_sha256": "9346b2508aa4e9def57598c282d6063e70eac35e5061c6c7440e765c9455472e", + "released_checkpoint_sha256": "76f89c544c803ddec130ca0eed6fe3ec554ca264d5e6102964dc17e5f05ff72e", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/soprano/epoch_0162.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/soprano.yaml", + "config_sha256": "be1905d757d50bfbdd0b2bea965c8b000c96e4995e1b5aafcd710a24f25048eb", + "inference_model_id": "specialists-jc-soprano-ep162", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-jc-tenor-ep119", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "tenor", + "epoch": 119, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/tenor/epoch_0119.pt", + "original_checkpoint_filename": "model_htdemucs_ep_119_si_sdr_11.1179.ckpt", + "original_checkpoint_sha256": "1c20f2a931c8def9e8622190984de3ed599c5913566441cea491c691e4a7f70a", + "released_checkpoint_sha256": "0c136befb105df1f6d89e70291981e0192e2f9a05d963febef2f0e58f36a243e", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/tenor/epoch_0119.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/tenor.yaml", + "config_sha256": "762562c7e9d24d51e5816bf722edad5aa7bdb0cd45901419077a1c68347951f3", + "inference_model_id": "specialists-jc-tenor-ep119", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-jc-vp-ep45", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "vp", + "epoch": 45, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/vp/epoch_0045.pt", + "original_checkpoint_filename": "model_htdemucs_ep_45_si_sdr_15.9240.ckpt", + "original_checkpoint_sha256": "931aae478a2aaceb0572dff59a7877987b67a58f78f07621e195cf696fff6dff", + "released_checkpoint_sha256": "1c90231919c3c9e2a1938007ea386b3c6e41c22389a777a285e724cd73f5bc4b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/vp/epoch_0045.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/vp.yaml", + "config_sha256": "8a95fc90f7f87c354b3e3c2017c8631e088dd223ee15603697f1a3a85ae60ef4", + "inference_model_id": "specialists-jc-vp-ep45", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-sk-alto-ep237", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "alto", + "epoch": 237, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/alto/epoch_0237.pt", + "original_checkpoint_filename": "model_htdemucs_ep_237_si_sdr_12.8485.ckpt", + "original_checkpoint_sha256": "133008dc46d3ebe3ef660c52c65e1252f9ff243a048f97a95d68b04a123ca6a4", + "released_checkpoint_sha256": "78a1673de410ec40b450a766c593f696796fc126259f324d62afe11ee62c6279", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/alto/epoch_0237.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/alto.yaml", + "config_sha256": "a0206c56e8078c7c3443eee4ba343f35baae36975ec403052bb99d0345c1e8b4", + "inference_model_id": "specialists-sk-alto-ep237", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-sk-bass-ep182", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "bass", + "epoch": 182, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/bass/epoch_0182.pt", + "original_checkpoint_filename": "model_htdemucs_ep_182_si_sdr_13.8862.ckpt", + "original_checkpoint_sha256": "9b24ac03c73f79810228f0dde23f430b5221f019c5dea7f30500708f4dfd5ac6", + "released_checkpoint_sha256": "07247e6587d36d4261be4fceda9d7bf1e401a3060935644222525abfe7b8ec34", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/bass/epoch_0182.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/bass.yaml", + "config_sha256": "4f0a1834931f5e2309e28b1d74059b1a77fff3d018037ddfdab7dcd09328c698", + "inference_model_id": "specialists-sk-bass-ep182", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-sk-baritone-ep43", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "baritone", + "epoch": 43, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/baritone/epoch_0043.pt", + "original_checkpoint_filename": "model_htdemucs_ep_43_si_sdr_11.1058.ckpt", + "original_checkpoint_sha256": "ffe9dafeff2c96595f479a406e327172c95355bed401d233410e9565c806d2ee", + "released_checkpoint_sha256": "f6affa1758e913c8ac5d9f2b8f6e7d505f6cda4efdc0aa76d5beea2a45319cd6", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/baritone/epoch_0043.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/baritone.yaml", + "config_sha256": "96464a49ea78eab9975e08392065b35e6da838a568d29c91dd15796bd39d03fe", + "inference_model_id": "specialists-sk-baritone-ep43", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-sk-soprano-ep220", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "soprano", + "epoch": 220, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/soprano/epoch_0220.pt", + "original_checkpoint_filename": "model_htdemucs_ep_220_si_sdr_12.3675.ckpt", + "original_checkpoint_sha256": "b9e4c1f9476f13ce673ea007756aec990318ccb4d571c8060355e9f1eccfab60", + "released_checkpoint_sha256": "474bdacd5cbf3df65c2b3153688304372ed6fe352658735b3cfc677a52e5e7ff", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/soprano/epoch_0220.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/soprano.yaml", + "config_sha256": "be1905d757d50bfbdd0b2bea965c8b000c96e4995e1b5aafcd710a24f25048eb", + "inference_model_id": "specialists-sk-soprano-ep220", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-sk-tenor-ep214", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "tenor", + "epoch": 214, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/tenor/epoch_0214.pt", + "original_checkpoint_filename": "model_htdemucs_ep_214_si_sdr_12.0113.ckpt", + "original_checkpoint_sha256": "2f3e5de523bb4bf8eb8fb77e25ce2d2962ac4b00e56b85b2e89614e049878005", + "released_checkpoint_sha256": "919f1808e8faa03c582accacc11913a0443dde4d32f02092ae302588aa7fbcca", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/tenor/epoch_0214.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/tenor.yaml", + "config_sha256": "762562c7e9d24d51e5816bf722edad5aa7bdb0cd45901419077a1c68347951f3", + "inference_model_id": "specialists-sk-tenor-ep214", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "specialists-sk-vp-ep205", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "vp", + "epoch": 205, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/vp/epoch_0205.pt", + "original_checkpoint_filename": "model_htdemucs_ep_205_si_sdr_13.7446.ckpt", + "original_checkpoint_sha256": "924efa1dd19fc793c0c8ffb32d0bd736a9791f01bc77f18d2f1eeedf06b7411d", + "released_checkpoint_sha256": "9f850e35bad3b83e8330e31b30f4a4f38460db9d02f3a3a798c3c4bc6294c77b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/vp/epoch_0205.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/vp.yaml", + "config_sha256": "8a95fc90f7f87c354b3e3c2017c8631e088dd223ee15603697f1a3a85ae60ef4", + "inference_model_id": "specialists-sk-vp-ep205", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k1-step0198000", + "family": "conditioned", + "research_id": "K1", + "description": "HTDemucs | vocal-part-conditioned | sk-natural | no-encoder | step 198000 (K1)", + "conditioning": "part", + "training_data": "sk-natural", + "encoder": "no-encoder", + "step": 198000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_shared" + ], + "paper_tables": [ + "cross_corpus" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt", + "original_checkpoint_sha256": "fbd19ff91075877839bb76b022d1b5ad018a980a26603ea9cb4cca7aac0feb6a", + "original_checkpoint_filename": "best_step0198000.pt", + "released_checkpoint_sha256": "9c59bfca391faf96e6d3f130d27dd75f4ed5f0eb1b950ed75361b343e926af14", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0198000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "supersedes": { + "step": 184000, + "sha256": "261daa7899501229143136c46a501097caba9ac903650bc9c475c3ee049407b0", + "preserve_historical": true + }, + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K1.json", + "config_sha256": "a2c358f669574dafa25403b5d6e4a0ee0fc3cf237e03399886761f95fd7c0fbd", + "inference_model_id": "Part_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "l1-step0141000", + "family": "conditioned", + "research_id": "L1", + "description": "HTDemucs | vocal-part-conditioned | jc-natural | no-encoder | step 141000 (L1)", + "conditioning": "part", + "training_data": "jc-natural", + "encoder": "no-encoder", + "step": 141000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_shared" + ], + "paper_tables": [ + "cross_corpus" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt", + "original_checkpoint_sha256": "92e19968bef2ce84d3800141afc50cc47141bd016adfcc07b70e705df7408004", + "original_checkpoint_filename": "best_step0141000.pt", + "released_checkpoint_sha256": "f9fdca2de1f89b650b603d386f39128b4b06b0ebd4a59469f57ce256e8c26065", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0141000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L1.json", + "config_sha256": "45fca96de4f1849d076368538dd789f97e5e0a4ab4443c6c0922e4505c2ac08a", + "inference_model_id": "Part_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k1-step0004000", + "family": "conditioned", + "research_id": "K1", + "description": "HTDemucs | vocal-part-conditioned | sk-natural | no-encoder | step 4000 (K1)", + "conditioning": "part", + "training_data": "sk-natural", + "encoder": "no-encoder", + "step": 4000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt", + "original_checkpoint_sha256": "979ba17521f398fb1b2a3e0ccad29615999b50e87d76347b2682212a61b49427", + "original_checkpoint_filename": "best_step0004000.pt", + "released_checkpoint_sha256": "31821cfe956fc6a7527e718d6702bcd71b84a8881484c1339753bca7207b316b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-sk-natural/sisdr-v1/no-encoder/step-0004000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K1.json", + "config_sha256": "a2c358f669574dafa25403b5d6e4a0ee0fc3cf237e03399886761f95fd7c0fbd", + "inference_model_id": "Part_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k5-step0012000", + "family": "conditioned", + "research_id": "K5", + "description": "HTDemucs | singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 12000 (K5)", + "conditioning": "singer", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 12000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_shared", + "jc_inclusive" + ], + "paper_tables": [ + "development_a", + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt", + "original_checkpoint_sha256": "6f8e199008a542904330495dab28d7f56d8c380a6959ea4e460b22450c56ca43", + "original_checkpoint_filename": "best_step0012000.pt", + "released_checkpoint_sha256": "89c5a8829aca66cdbb0033338dc83d769e06f98300f68464a666b87f73e80abf", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0012000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K5.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "inference_model_id": "Singer_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k7-step0007000", + "family": "conditioned", + "research_id": "K7", + "description": "HTDemucs | vocal-part + singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 7000 (K7)", + "conditioning": "dual", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 7000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt", + "original_checkpoint_sha256": "0646527189b1221552c01c918cff01173e98077dc515a6fe9f795ffab00c24ee", + "original_checkpoint_filename": "best_step0007000.pt", + "released_checkpoint_sha256": "528d75380b58ef43e266a0c1f6f3cd86883710d2a95ff9c492724790193fa6ab", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/dual/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0007000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K7.json", + "config_sha256": "5cbc1736ca73762cd34603e55cc3850e2cbf2872388898f710a4ff3e66b117dc", + "inference_model_id": "Dual_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161978465, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "l1-step0119000", + "family": "conditioned", + "research_id": "L1", + "description": "HTDemucs | vocal-part-conditioned | jc-natural | no-encoder | step 119000 (L1)", + "conditioning": "part", + "training_data": "jc-natural", + "encoder": "no-encoder", + "step": 119000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt", + "original_checkpoint_sha256": "5de579abaf11546eab1fcfa29566cf4934ece6c5fe7b23df56b6af27c3be3723", + "original_checkpoint_filename": "best_step0119000.pt", + "released_checkpoint_sha256": "6af03b62d8e5f419f72787b7909c3e82ad7d3194fa43c920a47a064374f63e8b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/role/train-jc-natural/sisdr-v1/no-encoder/step-0119000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L1.json", + "config_sha256": "45fca96de4f1849d076368538dd789f97e5e0a4ab4443c6c0922e4505c2ac08a", + "inference_model_id": "Part_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109889741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "l3-step0184000", + "family": "conditioned", + "research_id": "L3", + "description": "HTDemucs | singer-enrollment-conditioned | jc-natural | redimnet2-finetuned | step 184000 (L3)", + "conditioning": "singer", + "training_data": "jc-natural", + "encoder": "redimnet2-finetuned", + "step": 184000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt", + "original_checkpoint_sha256": "9f49b7d59575658dcfb9667395d998e8fc34a935b3995b099593e99ee1de5226", + "original_checkpoint_filename": "best_step0184000.pt", + "released_checkpoint_sha256": "7021317a39953018ea31d9f07578bbe4da9e3646ca62e1e59891abba57edd94d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0184000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L3.json", + "config_sha256": "a0bab6c22d17b9b23eca07c5cd3f2f41ffcb01d807ea78a19f3c131eccd91438", + "inference_model_id": "Singer_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "l4-step0188000", + "family": "conditioned", + "research_id": "L4", + "description": "HTDemucs | vocal-part + singer-enrollment-conditioned | jc-natural | redimnet2-finetuned | step 188000 (L4)", + "conditioning": "dual", + "training_data": "jc-natural", + "encoder": "redimnet2-finetuned", + "step": 188000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_inclusive" + ], + "paper_tables": [ + "target_conditioning" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt", + "original_checkpoint_sha256": "724af8d2906fc08de54b7484df69143a3a7b2c62c412e6e473b979bc34cd74d0", + "original_checkpoint_filename": "best_step0188000.pt", + "released_checkpoint_sha256": "511dc6729da436c5fe72960e9a1de4968578fbb985fd00db5eb653e2d605dfb9", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/dual/train-jc-natural/sisdr-v1/redimnet2-finetuned/step-0188000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/L4.json", + "config_sha256": "7c46503ca0ddc32623e184e801fe411184ff3a3d42dcddaf34106fb33d45f2cf", + "inference_model_id": "Dual_conditioned_htdemucs_JC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161978465, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k5-step0179000", + "family": "conditioned", + "research_id": "K5", + "description": "HTDemucs | singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 179000 (K5)", + "conditioning": "singer", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 179000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_shared" + ], + "paper_tables": [ + "development_a" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt", + "original_checkpoint_sha256": "eeeac981bacdd063207a0258fb60625edd2b60e94bc544852a1c2459223a04d3", + "original_checkpoint_filename": "best_step0179000.pt", + "released_checkpoint_sha256": "537531081da4c7c1a7d5568f8847bfaab1df3d54658486381bdd7f3d2f7f8d7d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0179000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K5.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "inference_model_id": "Singer_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k5-step0106000", + "family": "conditioned", + "research_id": "K5", + "description": "HTDemucs | singer-enrollment-conditioned | sk-natural | redimnet2-finetuned | step 106000 (K5)", + "conditioning": "singer", + "training_data": "sk-natural", + "encoder": "redimnet2-finetuned", + "step": 106000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "combined_shared", + "combined_inclusive" + ], + "paper_tables": [ + "development_a", + "development_b" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt", + "original_checkpoint_sha256": "df846b039ddd5e032f221adcf3de33a9a3f546f21b47f0bc3d30df891ab4ca93", + "original_checkpoint_filename": "best_step0106000.pt", + "released_checkpoint_sha256": "b28626e30a12ba43337b3cec3da2558685ddcc649fe34dd78e7e71590e922b29", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-natural/sisdr-v1/redimnet2-finetuned/step-0106000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/conditioned/K5.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "inference_model_id": "Singer_conditioned_htdemucs_SK", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "m3-step0198000", + "family": "conditioned", + "research_id": "M3", + "description": "HTDemucs | singer-enrollment-conditioned | jc-sk-natural | redimnet2-finetuned | step 198000 (M3)", + "conditioning": "singer", + "training_data": "jc-sk-natural", + "encoder": "redimnet2-finetuned", + "step": 198000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "sk_shared" + ], + "paper_tables": [ + "development_a" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt", + "original_checkpoint_sha256": "6997e980ed2dded55ba1db2d5e88b71ad7024782125b2bd8dcf41df71189d6b4", + "original_checkpoint_filename": "best_step0198000.pt", + "released_checkpoint_sha256": "67c68a7f12076db456166edab0120f94947806689eca50da546f7adac260d77b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0198000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "supersedes": { + "step": 159000, + "sha256": "03e74a6ae7f56d647c66e582f368f714b5504f8c657e78fb7f6c2142d0ae8ab1", + "preserve_historical": true + }, + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/M3.json", + "config_sha256": "fee6def3080111793b362b2b9fda724720ddd13f28b2596f2a3eeb95fb8168d0", + "inference_model_id": "Singer_conditioned_htdemucs_SKnJC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "m3-step0195000", + "family": "conditioned", + "research_id": "M3", + "description": "HTDemucs | singer-enrollment-conditioned | jc-sk-natural | redimnet2-finetuned | step 195000 (M3)", + "conditioning": "singer", + "training_data": "jc-sk-natural", + "encoder": "redimnet2-finetuned", + "step": 195000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "jc_shared", + "combined_shared" + ], + "paper_tables": [ + "development_a" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt", + "original_checkpoint_sha256": "badc60d5bb2f2cf0cca92b11dc98c5a4de51fcd41312fff43607f7664ba5f9df", + "original_checkpoint_filename": "best_step0195000.pt", + "released_checkpoint_sha256": "12a92b20a7b1fed9614c48b925eaeb2f563428201d5c270c11e3f7c3df7359e7", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-jc-sk-natural/sisdr-v1/redimnet2-finetuned/step-0195000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "supersedes": { + "step": 173000, + "sha256": "f99b7ee04a026c2c628bcfaac3ddbe7a5ffabc95bb96e9d0889e3079e8db1b06", + "preserve_historical": true + }, + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/conditioned/M3.json", + "config_sha256": "fee6def3080111793b362b2b9fda724720ddd13f28b2596f2a3eeb95fb8168d0", + "inference_model_id": "Singer_conditioned_htdemucs_SKnJC", + "artifact_format": "acappellaset-conditioned-inference-v1", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 161878741, + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k6-step0131000", + "family": "conditioned", + "research_id": "K6", + "description": "HTDemucs | singer-enrollment-conditioned | sk-full-remix | redimnet2-finetuned | step 131000 (K6)", + "conditioning": "singer", + "training_data": "sk-full-remix", + "encoder": "redimnet2-finetuned", + "step": 131000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "combined_inclusive" + ], + "paper_tables": [ + "development_b" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt", + "original_checkpoint_sha256": "ff74b497341edcfe6a6357749ed3e6bff795f1740ef022f4003f1241f198e517", + "original_checkpoint_filename": "best_step0131000.pt", + "released_checkpoint_sha256": "893326c96ac17b9cfed1cd5126176763f6a0a525115e3322c9a701ef232a9e69", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-full-remix/sisdr-v1/redimnet2-finetuned/step-0131000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "artifact_format": "acappellaset-conditioned-inference-v1", + "notice_path": "WEIGHTS_NOTICE.md", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "config": "configs/conditioned/Singer_conditioned_htdemucs_SK_full_remix.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "released_checkpoint_bytes": 161878741, + "inference_model_id": "Singer_conditioned_htdemucs_SK_full_remix", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + }, + { + "id": "k6-no-aug-step0048000", + "family": "conditioned", + "research_id": "K6_no_aug", + "description": "HTDemucs | singer-enrollment-conditioned | sk-single-clone | redimnet2-finetuned | step 48000 (K6_no_aug)", + "conditioning": "singer", + "training_data": "sk-single-clone", + "encoder": "redimnet2-finetuned", + "step": 48000, + "selection_horizon_updates": 200000, + "selection_labels": [ + "combined_inclusive" + ], + "paper_tables": [ + "development_b" + ], + "release_inclusion": "confirmed", + "artifact_status": "published", + "public_inference_status": "published_export_strict_loading_verified", + "planned_checkpoint_path": "conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt", + "original_checkpoint_sha256": "79ee3dd239925ea2c075641eadc0e13c001a81f9f367f8ee6881c5e8a3881901", + "original_checkpoint_filename": "best_step0048000.pt", + "released_checkpoint_sha256": "00fc378e77c7148f75e853ba8e84f246aba1557c30ec3e5f205e4227099c5c65", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/conditioned/enrollment/train-sk-single-clone/sisdr-v1/redimnet2-finetuned/step-0048000.pt", + "weight_license": "cc-by-nc-4.0", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "artifact_format": "acappellaset-conditioned-inference-v1", + "notice_path": "WEIGHTS_NOTICE.md", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "config": "configs/conditioned/Singer_conditioned_htdemucs_SK_single_clone.json", + "config_sha256": "fb94dcf47cd96b4ee07b0975996a300d9c0938e3a24b855cfbc273fe60f58c8a", + "released_checkpoint_bytes": 161878741, + "inference_model_id": "Singer_conditioned_htdemucs_SK_single_clone", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" + } + ], + "additional_models": "29 selections: 3 original fixed, 12 original mixture-only specialists, and 14 conditioned checkpoints from 9 recipes, including SK full-remix and SK single-clone augmentation.", + "repository_id": "AcappellaSet/ACappellaSet-Models", + "excluded_families": [ + { + "family": "later_singer_enrollment_specialists", + "decision_date": "2026-10-14", + "scope": "public_release", + "original_part_specific_specialists": "retained" + }, + { + "family": "newer_fixed_six_training_runs", + "decision_date": "2026-10-14", + "scope": "icassp_release", + "reason": "Not used in the supplied Results tables; AI-augmentation results use shared singer-conditioned models." + } + ], + "paper_checkpoint_policy": "Use final-200k validation selections with matching updated evaluations: K1 SK-shared198k; M3 JC/pooled-shared195k and SK-shared198k. Historical selections remain recorded under supersedes. Manuscript edits are separate from this metadata update.", + "project_checkpoint_count": 29, + "evaluation_scope": { + "development_a": "shared selection and shared evaluation", + "development_b": "pooled-inclusive selection and inclusive evaluation", + "shared_aggregation": "pair-weighted over retained S/A/T/B/VP pairs, not a macro-average of part means", + "selection_label_alias": "combined means pooled JC+SK validation; it does not indicate a training corpus" + }, + "published_checkpoint_count": 29, + "active_release_checkpoint_count": 29, + "deferred_checkpoint_count": 0, + "preferred_weight_license": "cc-by-nc-4.0", + "previous_repository_id": "pting126/ACappellaSet-models", + "repository_visibility": "private" +} diff --git a/specialists/JC/legacy/alto/epoch_0223.pt b/specialists/JC/legacy/alto/epoch_0223.pt new file mode 100644 index 0000000000000000000000000000000000000000..1ee133b11a0d375465e0162946b99f1022973d20 --- /dev/null +++ b/specialists/JC/legacy/alto/epoch_0223.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9212124f4a3eab2b9a4a14669bc3e52c66de3b1b795e805a5b575e430f24072d +size 109785182 diff --git a/specialists/JC/legacy/alto/model.json b/specialists/JC/legacy/alto/model.json new file mode 100644 index 0000000000000000000000000000000000000000..501e870926c68838ce72a842d21fd1e8fee8df0e --- /dev/null +++ b/specialists/JC/legacy/alto/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-jc-alto-ep223", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "alto", + "epoch": 223, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/alto/epoch_0223.pt", + "original_checkpoint_filename": "model_htdemucs_ep_223_si_sdr_6.0926.ckpt", + "original_checkpoint_sha256": "5c8118424cbe5264b953f37d8336194b66089bac0aa18c1a32d1c41b386e7e1e", + "released_checkpoint_sha256": "9212124f4a3eab2b9a4a14669bc3e52c66de3b1b795e805a5b575e430f24072d", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/alto/epoch_0223.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/alto.yaml", + "config_sha256": "a0206c56e8078c7c3443eee4ba343f35baae36975ec403052bb99d0345c1e8b4", + "inference_model_id": "specialists-jc-alto-ep223", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/JC/legacy/bank.json b/specialists/JC/legacy/bank.json new file mode 100644 index 0000000000000000000000000000000000000000..ba49abeaed3a42d6936308f206ef41ca76823f8b --- /dev/null +++ b/specialists/JC/legacy/bank.json @@ -0,0 +1,14 @@ +{ + "training_corpus": "JC", + "recipe": "legacy", + "enrollment_required": false, + "artifact_status": "published", + "roles": { + "alto": "specialists-jc-alto-ep223", + "bass": "specialists-jc-bass-ep110", + "lead": "specialists-jc-lead-ep149", + "soprano": "specialists-jc-soprano-ep162", + "tenor": "specialists-jc-tenor-ep119", + "vp": "specialists-jc-vp-ep45" + } +} diff --git a/specialists/JC/legacy/bass/epoch_0110.pt b/specialists/JC/legacy/bass/epoch_0110.pt new file mode 100644 index 0000000000000000000000000000000000000000..19201463716485b88e39e6f9c06331f2a2e5401f --- /dev/null +++ b/specialists/JC/legacy/bass/epoch_0110.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b4b70dd70fa921cfe7d5b79c8e31004c5ea9893a9de659b2f5592fcde4a213c +size 109785182 diff --git a/specialists/JC/legacy/bass/model.json b/specialists/JC/legacy/bass/model.json new file mode 100644 index 0000000000000000000000000000000000000000..f9cc3e704116abeb7b395553368ad353e5a60c81 --- /dev/null +++ b/specialists/JC/legacy/bass/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-jc-bass-ep110", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "bass", + "epoch": 110, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/bass/epoch_0110.pt", + "original_checkpoint_filename": "model_htdemucs_ep_110_si_sdr_16.6663.ckpt", + "original_checkpoint_sha256": "23d152aa2ae0250fe852ac5e31ae68f2d70736acd4cb2a5f9515a9b10302a810", + "released_checkpoint_sha256": "7b4b70dd70fa921cfe7d5b79c8e31004c5ea9893a9de659b2f5592fcde4a213c", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/bass/epoch_0110.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/bass.yaml", + "config_sha256": "4f0a1834931f5e2309e28b1d74059b1a77fff3d018037ddfdab7dcd09328c698", + "inference_model_id": "specialists-jc-bass-ep110", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/JC/legacy/lead/epoch_0149.pt b/specialists/JC/legacy/lead/epoch_0149.pt new file mode 100644 index 0000000000000000000000000000000000000000..d3906800bf7f5ed22f2c91a2d52d1176e4c553d2 --- /dev/null +++ b/specialists/JC/legacy/lead/epoch_0149.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4825c5864e9297d2e1bc7977ea5879af522f182b56483581bf14b90b8c18f465 +size 109785182 diff --git a/specialists/JC/legacy/lead/model.json b/specialists/JC/legacy/lead/model.json new file mode 100644 index 0000000000000000000000000000000000000000..bc6a89b5a5875302d5068e6ce980205151683e42 --- /dev/null +++ b/specialists/JC/legacy/lead/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-jc-lead-ep149", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "lead", + "epoch": 149, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/lead/epoch_0149.pt", + "original_checkpoint_filename": "model_htdemucs_ep_149_si_sdr_9.6743.ckpt", + "original_checkpoint_sha256": "0a37e456852527fd2d0c7d983ba086f1ebb15064a39633f99c06cbbb37914c28", + "released_checkpoint_sha256": "4825c5864e9297d2e1bc7977ea5879af522f182b56483581bf14b90b8c18f465", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/lead/epoch_0149.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/lead_vocal.yaml", + "config_sha256": "dcb7bb687bad3a7f00d03491b3fba10fa12a2b5777025cc04c2a68fcfda311e0", + "inference_model_id": "specialists-jc-lead-ep149", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/JC/legacy/soprano/epoch_0162.pt b/specialists/JC/legacy/soprano/epoch_0162.pt new file mode 100644 index 0000000000000000000000000000000000000000..ef9445401496a98e0caffb223b8145e2c2ff2ee8 --- /dev/null +++ b/specialists/JC/legacy/soprano/epoch_0162.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76f89c544c803ddec130ca0eed6fe3ec554ca264d5e6102964dc17e5f05ff72e +size 109785182 diff --git a/specialists/JC/legacy/soprano/model.json b/specialists/JC/legacy/soprano/model.json new file mode 100644 index 0000000000000000000000000000000000000000..7c814cb37bb7dccf151724771962030622ef48dc --- /dev/null +++ b/specialists/JC/legacy/soprano/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-jc-soprano-ep162", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "soprano", + "epoch": 162, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/soprano/epoch_0162.pt", + "original_checkpoint_filename": "model_htdemucs_ep_162_si_sdr_6.1072.ckpt", + "original_checkpoint_sha256": "9346b2508aa4e9def57598c282d6063e70eac35e5061c6c7440e765c9455472e", + "released_checkpoint_sha256": "76f89c544c803ddec130ca0eed6fe3ec554ca264d5e6102964dc17e5f05ff72e", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/soprano/epoch_0162.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/soprano.yaml", + "config_sha256": "be1905d757d50bfbdd0b2bea965c8b000c96e4995e1b5aafcd710a24f25048eb", + "inference_model_id": "specialists-jc-soprano-ep162", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/JC/legacy/tenor/epoch_0119.pt b/specialists/JC/legacy/tenor/epoch_0119.pt new file mode 100644 index 0000000000000000000000000000000000000000..3836907ccc61df1a0e69bfd653256fad416b7b0d --- /dev/null +++ b/specialists/JC/legacy/tenor/epoch_0119.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c136befb105df1f6d89e70291981e0192e2f9a05d963febef2f0e58f36a243e +size 109785182 diff --git a/specialists/JC/legacy/tenor/model.json b/specialists/JC/legacy/tenor/model.json new file mode 100644 index 0000000000000000000000000000000000000000..92d7bb039850665c2a356071473e557a93899998 --- /dev/null +++ b/specialists/JC/legacy/tenor/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-jc-tenor-ep119", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "tenor", + "epoch": 119, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/tenor/epoch_0119.pt", + "original_checkpoint_filename": "model_htdemucs_ep_119_si_sdr_11.1179.ckpt", + "original_checkpoint_sha256": "1c20f2a931c8def9e8622190984de3ed599c5913566441cea491c691e4a7f70a", + "released_checkpoint_sha256": "0c136befb105df1f6d89e70291981e0192e2f9a05d963febef2f0e58f36a243e", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/tenor/epoch_0119.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/tenor.yaml", + "config_sha256": "762562c7e9d24d51e5816bf722edad5aa7bdb0cd45901419077a1c68347951f3", + "inference_model_id": "specialists-jc-tenor-ep119", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/JC/legacy/vp/epoch_0045.pt b/specialists/JC/legacy/vp/epoch_0045.pt new file mode 100644 index 0000000000000000000000000000000000000000..46128d04c6b3e5dcf535438d60f525edf227120c --- /dev/null +++ b/specialists/JC/legacy/vp/epoch_0045.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1c90231919c3c9e2a1938007ea386b3c6e41c22389a777a285e724cd73f5bc4b +size 109785182 diff --git a/specialists/JC/legacy/vp/model.json b/specialists/JC/legacy/vp/model.json new file mode 100644 index 0000000000000000000000000000000000000000..e19bcea61fc03f07e6594bdca58c6d1a8639bdea --- /dev/null +++ b/specialists/JC/legacy/vp/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-jc-vp-ep45", + "family": "specialists", + "training_corpus": "JC", + "recipe": "legacy", + "target_role": "vp", + "epoch": 45, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/JC/legacy/vp/epoch_0045.pt", + "original_checkpoint_filename": "model_htdemucs_ep_45_si_sdr_15.9240.ckpt", + "original_checkpoint_sha256": "931aae478a2aaceb0572dff59a7877987b67a58f78f07621e195cf696fff6dff", + "released_checkpoint_sha256": "1c90231919c3c9e2a1938007ea386b3c6e41c22389a777a285e724cd73f5bc4b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/JC/legacy/vp/epoch_0045.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "license_review_status": "assigned; applicable upstream notices preserved", + "notice_path": "WEIGHTS_NOTICE.md", + "config": "configs/specialists/JC/vp.yaml", + "config_sha256": "8a95fc90f7f87c354b3e3c2017c8631e088dd223ee15603697f1a3a85ae60ef4", + "inference_model_id": "specialists-jc-vp-ep45", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/README.md b/specialists/README.md new file mode 100644 index 0000000000000000000000000000000000000000..96f05e039b967945115b146093f8c63c9d46eca7 --- /dev/null +++ b/specialists/README.md @@ -0,0 +1,6 @@ +# Part-specific specialists + +Confirmed initial release: six historical JC specialists and six historical SK specialists. Each bank has one model per target part. These historical models do not require enrollment audio. Checkpoint paths and SHA-256 hashes are listed in [manifest.json](../manifest.json); weight license: CC BY-NC 4.0. + +The later singer-enrollment specialist experiments are excluded from the public +release. The [public inference code](https://github.com/pting126/ACappellaSet/blob/main/docs/inference.md#original-fixed-and-part-specific-models) supports these original part-specific banks with verified local checkpoints. Short-input CPU parity against the native implementation has been verified; the inference-only exports preserve every model tensor and omit optimizer state. diff --git a/specialists/SK/legacy/alto/epoch_0237.pt b/specialists/SK/legacy/alto/epoch_0237.pt new file mode 100644 index 0000000000000000000000000000000000000000..5cbe883ee6cb57f740a592af2c8e7c6efc029c15 --- /dev/null +++ b/specialists/SK/legacy/alto/epoch_0237.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78a1673de410ec40b450a766c593f696796fc126259f324d62afe11ee62c6279 +size 109785182 diff --git a/specialists/SK/legacy/alto/model.json b/specialists/SK/legacy/alto/model.json new file mode 100644 index 0000000000000000000000000000000000000000..f9762ca34653f548ea987e6959eb43005f3f856a --- /dev/null +++ b/specialists/SK/legacy/alto/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-sk-alto-ep237", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "alto", + "epoch": 237, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/alto/epoch_0237.pt", + "original_checkpoint_filename": "model_htdemucs_ep_237_si_sdr_12.8485.ckpt", + "original_checkpoint_sha256": "133008dc46d3ebe3ef660c52c65e1252f9ff243a048f97a95d68b04a123ca6a4", + "released_checkpoint_sha256": "78a1673de410ec40b450a766c593f696796fc126259f324d62afe11ee62c6279", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/alto/epoch_0237.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/alto.yaml", + "config_sha256": "a0206c56e8078c7c3443eee4ba343f35baae36975ec403052bb99d0345c1e8b4", + "inference_model_id": "specialists-sk-alto-ep237", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/SK/legacy/bank.json b/specialists/SK/legacy/bank.json new file mode 100644 index 0000000000000000000000000000000000000000..95abb7ffd84facfe0c0dde59bffdb25edbe5d268 --- /dev/null +++ b/specialists/SK/legacy/bank.json @@ -0,0 +1,14 @@ +{ + "training_corpus": "SK", + "recipe": "legacy", + "enrollment_required": false, + "artifact_status": "published", + "roles": { + "alto": "specialists-sk-alto-ep237", + "bass": "specialists-sk-bass-ep182", + "baritone": "specialists-sk-baritone-ep43", + "soprano": "specialists-sk-soprano-ep220", + "tenor": "specialists-sk-tenor-ep214", + "vp": "specialists-sk-vp-ep205" + } +} diff --git a/specialists/SK/legacy/baritone/epoch_0043.pt b/specialists/SK/legacy/baritone/epoch_0043.pt new file mode 100644 index 0000000000000000000000000000000000000000..c07e16de339045c819f48c9cb48ada10fcaf40f9 --- /dev/null +++ b/specialists/SK/legacy/baritone/epoch_0043.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f6affa1758e913c8ac5d9f2b8f6e7d505f6cda4efdc0aa76d5beea2a45319cd6 +size 109785182 diff --git a/specialists/SK/legacy/baritone/model.json b/specialists/SK/legacy/baritone/model.json new file mode 100644 index 0000000000000000000000000000000000000000..cdb925deb52a42aa6d21af4df9e21d0d5b5cb2eb --- /dev/null +++ b/specialists/SK/legacy/baritone/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-sk-baritone-ep43", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "baritone", + "epoch": 43, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/baritone/epoch_0043.pt", + "original_checkpoint_filename": "model_htdemucs_ep_43_si_sdr_11.1058.ckpt", + "original_checkpoint_sha256": "ffe9dafeff2c96595f479a406e327172c95355bed401d233410e9565c806d2ee", + "released_checkpoint_sha256": "f6affa1758e913c8ac5d9f2b8f6e7d505f6cda4efdc0aa76d5beea2a45319cd6", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/baritone/epoch_0043.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/baritone.yaml", + "config_sha256": "96464a49ea78eab9975e08392065b35e6da838a568d29c91dd15796bd39d03fe", + "inference_model_id": "specialists-sk-baritone-ep43", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/SK/legacy/bass/epoch_0182.pt b/specialists/SK/legacy/bass/epoch_0182.pt new file mode 100644 index 0000000000000000000000000000000000000000..370d6d35489d213424f45d8942978292b1059deb --- /dev/null +++ b/specialists/SK/legacy/bass/epoch_0182.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07247e6587d36d4261be4fceda9d7bf1e401a3060935644222525abfe7b8ec34 +size 109785182 diff --git a/specialists/SK/legacy/bass/model.json b/specialists/SK/legacy/bass/model.json new file mode 100644 index 0000000000000000000000000000000000000000..83f584de1eb8348b3fdecd2691a85658e6c78a7c --- /dev/null +++ b/specialists/SK/legacy/bass/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-sk-bass-ep182", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "bass", + "epoch": 182, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/bass/epoch_0182.pt", + "original_checkpoint_filename": "model_htdemucs_ep_182_si_sdr_13.8862.ckpt", + "original_checkpoint_sha256": "9b24ac03c73f79810228f0dde23f430b5221f019c5dea7f30500708f4dfd5ac6", + "released_checkpoint_sha256": "07247e6587d36d4261be4fceda9d7bf1e401a3060935644222525abfe7b8ec34", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/bass/epoch_0182.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/bass.yaml", + "config_sha256": "4f0a1834931f5e2309e28b1d74059b1a77fff3d018037ddfdab7dcd09328c698", + "inference_model_id": "specialists-sk-bass-ep182", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/SK/legacy/soprano/epoch_0220.pt b/specialists/SK/legacy/soprano/epoch_0220.pt new file mode 100644 index 0000000000000000000000000000000000000000..56749d6dbfd14cdf343da058202f0e5ae397d8e6 --- /dev/null +++ b/specialists/SK/legacy/soprano/epoch_0220.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:474bdacd5cbf3df65c2b3153688304372ed6fe352658735b3cfc677a52e5e7ff +size 109785182 diff --git a/specialists/SK/legacy/soprano/model.json b/specialists/SK/legacy/soprano/model.json new file mode 100644 index 0000000000000000000000000000000000000000..8100200b06cf4d7d232b1de2ad23c5239f477333 --- /dev/null +++ b/specialists/SK/legacy/soprano/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-sk-soprano-ep220", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "soprano", + "epoch": 220, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/soprano/epoch_0220.pt", + "original_checkpoint_filename": "model_htdemucs_ep_220_si_sdr_12.3675.ckpt", + "original_checkpoint_sha256": "b9e4c1f9476f13ce673ea007756aec990318ccb4d571c8060355e9f1eccfab60", + "released_checkpoint_sha256": "474bdacd5cbf3df65c2b3153688304372ed6fe352658735b3cfc677a52e5e7ff", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/soprano/epoch_0220.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/soprano.yaml", + "config_sha256": "be1905d757d50bfbdd0b2bea965c8b000c96e4995e1b5aafcd710a24f25048eb", + "inference_model_id": "specialists-sk-soprano-ep220", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/SK/legacy/tenor/epoch_0214.pt b/specialists/SK/legacy/tenor/epoch_0214.pt new file mode 100644 index 0000000000000000000000000000000000000000..dbf6be41884717fc1e0e9454eb25096b44581685 --- /dev/null +++ b/specialists/SK/legacy/tenor/epoch_0214.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:919f1808e8faa03c582accacc11913a0443dde4d32f02092ae302588aa7fbcca +size 109785182 diff --git a/specialists/SK/legacy/tenor/model.json b/specialists/SK/legacy/tenor/model.json new file mode 100644 index 0000000000000000000000000000000000000000..50f158069964747c7d0ee4e17faca210028533ea --- /dev/null +++ b/specialists/SK/legacy/tenor/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-sk-tenor-ep214", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "tenor", + "epoch": 214, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/tenor/epoch_0214.pt", + "original_checkpoint_filename": "model_htdemucs_ep_214_si_sdr_12.0113.ckpt", + "original_checkpoint_sha256": "2f3e5de523bb4bf8eb8fb77e25ce2d2962ac4b00e56b85b2e89614e049878005", + "released_checkpoint_sha256": "919f1808e8faa03c582accacc11913a0443dde4d32f02092ae302588aa7fbcca", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/tenor/epoch_0214.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/tenor.yaml", + "config_sha256": "762562c7e9d24d51e5816bf722edad5aa7bdb0cd45901419077a1c68347951f3", + "inference_model_id": "specialists-sk-tenor-ep214", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +} diff --git a/specialists/SK/legacy/vp/epoch_0205.pt b/specialists/SK/legacy/vp/epoch_0205.pt new file mode 100644 index 0000000000000000000000000000000000000000..a5f224b62dbd1371cb59c7ee0a18b89b6ad5f413 --- /dev/null +++ b/specialists/SK/legacy/vp/epoch_0205.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9f850e35bad3b83e8330e31b30f4a4f38460db9d02f3a3a798c3c4bc6294c77b +size 109785182 diff --git a/specialists/SK/legacy/vp/model.json b/specialists/SK/legacy/vp/model.json new file mode 100644 index 0000000000000000000000000000000000000000..ba7511d6d5aa3a4d724e877d588adfc93f68b35a --- /dev/null +++ b/specialists/SK/legacy/vp/model.json @@ -0,0 +1,33 @@ +{ + "id": "specialists-sk-vp-ep205", + "family": "specialists", + "training_corpus": "SK", + "recipe": "legacy", + "target_role": "vp", + "epoch": 205, + "release_inclusion": "confirmed", + "artifact_status": "published", + "planned_checkpoint_path": "specialists/SK/legacy/vp/epoch_0205.pt", + "original_checkpoint_filename": "model_htdemucs_ep_205_si_sdr_13.7446.ckpt", + "original_checkpoint_sha256": "924efa1dd19fc793c0c8ffb32d0bd736a9791f01bc77f18d2f1eeedf06b7411d", + "released_checkpoint_sha256": "9f850e35bad3b83e8330e31b30f4a4f38460db9d02f3a3a798c3c4bc6294c77b", + "download_url": "https://huggingface.co/AcappellaSet/ACappellaSet-Models/resolve/main/specialists/SK/legacy/vp/epoch_0205.pt", + "weight_license": "cc-by-nc-4.0", + "output_contract": { + "residual_index": 0, + "target_index": 1 + }, + "artifact_format": "pytorch_state_dict", + "original_hash_verification": "PASS, full-file SHA-256, 2026-10-14", + "intended_weight_license": "cc-by-nc-4.0", + "notice_path": "WEIGHTS_NOTICE.md", + "license_review_status": "assigned; applicable upstream notices preserved", + "config": "configs/specialists/SK/vp.yaml", + "config_sha256": "8a95fc90f7f87c354b3e3c2017c8631e088dd223ee15603697f1a3a85ae60ef4", + "inference_model_id": "specialists-sk-vp-ep205", + "checkpoint_contents": "Inference-only checkpoint; all original model and matching encoder tensors preserved; no optimizer state, training arguments, or internal paths.", + "released_checkpoint_bytes": 109785182, + "public_inference_status": "published_export_strict_loading_verified", + "weights_only": true, + "validation_report": "https://github.com/pting126/ACappellaSet/blob/main/docs/validation/inference-exports-2026-10-22.json" +}