Seed-VC (models)

models/Seed-VC (memescreamer)/.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

models/Seed-VC (memescreamer)/DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ema.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4145da6704ca0950c2ef6312c1cd0c15c370b13613faed2b4a9d9d71bb4ed036
+size 820863652

models/Seed-VC (memescreamer)/DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:883d19be9fe9c4bb2dce30d545cd77e5ed2b6326c7ed1c4d9a7ce8e555b163f9
+size 820864573

models/Seed-VC (memescreamer)/DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema_v2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42aef93ffe65857c840d270252fa040f7ba04514945ec460f3ac1ac2a96de684
+size 820865494

models/Seed-VC (memescreamer)/DiT_seed_v2_uvit_whisper_small_wavenet_bigvgan_pruned.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ec8841b20bb46df9f7e8e570a6946a4b87b940133c7f0e778487ff33841f720
+size 440312082

models/Seed-VC (memescreamer)/DiT_uvit_tat_xlsr_ema.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c853ea578b409f625f961bcb15d5cff1f8ef9a75f3209ec21d9b7c73ab422e88
+size 142112203

models/Seed-VC (memescreamer)/README.md

@@ -0,0 +1,18 @@
+---
+tags:
+  - mirror
+  - raidio-bot
+---
+
+# Mirror of Plachta/Seed-VC
+
+This is a pinned mirror of [Plachta/Seed-VC](https://huggingface.co/Plachta/Seed-VC).
+
+| Field | Value |
+|-------|-------|
+| Upstream | [Plachta/Seed-VC](https://huggingface.co/Plachta/Seed-VC) |
+| Revision | `257283f9f41585055e8f858fba4fd044e5caed6e` |
+| Mirror org | [memescreamer](https://huggingface.co/memescreamer) |
+
+This mirror exists to ensure availability of pinned model weights
+regardless of upstream changes.

models/Seed-VC (memescreamer)/config_dit_mel_seed_uvit_whisper_base_f0_44k.yml

@@ -0,0 +1,98 @@
+log_dir: "./runs/run_dit_mel_seed_uvit_whisper_base_f0_44k"
+save_freq: 1
+log_interval: 10
+save_interval: 1000
+device: "cuda"
+epochs: 1000 # number of epochs for first stage training (pre-training)
+batch_size: 1
+batch_length: 100 # maximum duration of audio in a batch (in seconds)
+max_len: 80 # maximum number of frames
+pretrained_model: ""
+pretrained_encoder: ""
+load_only_params: False # set to true if do not want to load epoch numbers and optimizer parameters
+
+preprocess_params:
+  sr: 44100
+  spect_params:
+    n_fft: 2048
+    win_length: 2048
+    hop_length: 512
+    n_mels: 128
+    fmin: 0
+    fmax: "None"
+
+model_params:
+  dit_type: "DiT" # uDiT or DiT
+  reg_loss_type: "l1" # l1 or l2
+
+  timbre_shifter:
+    se_db_path: "./modules/openvoice/checkpoints_v2/converter/se_db.pt"
+    ckpt_path: './modules/openvoice/checkpoints_v2/converter'
+
+  vocoder:
+    type: "bigvgan"
+    name: "nvidia/bigvgan_v2_44khz_128band_512x"
+
+  speech_tokenizer:
+    type: 'whisper'
+    name: "openai/whisper-small"
+
+  style_encoder:
+    dim: 192
+    campplus_path: "campplus_cn_common.bin"
+
+  DAC:
+    encoder_dim: 64
+    encoder_rates: [2, 5, 5, 6]
+    decoder_dim: 1536
+    decoder_rates: [ 6, 5, 5, 2 ]
+    sr: 24000
+
+  length_regulator:
+    channels: 768
+    is_discrete: false
+    in_channels: 768
+    content_codebook_size: 2048
+    sampling_ratios: [1, 1, 1, 1]
+    vector_quantize: false
+    n_codebooks: 1
+    quantizer_dropout: 0.0
+    f0_condition: true
+    n_f0_bins: 256
+
+  DiT:
+    hidden_dim: 768
+    num_heads: 12
+    depth: 17
+    class_dropout_prob: 0.1
+    block_size: 8192
+    in_channels: 128
+    style_condition: true
+    final_layer_type: 'mlp'
+    target: 'mel' # mel or codec
+    content_dim: 768
+    content_codebook_size: 1024
+    content_type: 'discrete'
+    f0_condition: true
+    n_f0_bins: 256
+    content_codebooks: 1
+    is_causal: false
+    long_skip_connection: false
+    zero_prompt_speech_token: false # for prompt component, do not input corresponding speech token
+    time_as_token: false
+    style_as_token: false
+    uvit_skip_connection: true
+    add_resblock_in_transformer: false
+
+  wavenet:
+    hidden_dim: 768
+    num_layers: 8
+    kernel_size: 5
+    dilation_rate: 1
+    p_dropout: 0.2
+    style_condition: true
+
+loss_params:
+  base_lr: 0.0001
+  lambda_mel: 45
+  lambda_kl: 1.0

models/Seed-VC (memescreamer)/config_dit_mel_seed_uvit_whisper_small_wavenet.yml

@@ -0,0 +1,91 @@
+log_dir: "./runs/run_dit_mel_seed_uvit_whisper_small_wavenet"
+save_freq: 1
+log_interval: 10
+save_interval: 1000
+device: "cuda"
+epochs: 1000 # number of epochs for first stage training (pre-training)
+batch_size: 2
+batch_length: 100 # maximum duration of audio in a batch (in seconds)
+max_len: 80 # maximum number of frames
+pretrained_model: ""
+pretrained_encoder: "./temp_ckpt.pth"
+load_only_params: False # set to true if do not want to load epoch numbers and optimizer parameters
+
+preprocess_params:
+  sr: 22050
+  spect_params:
+    n_fft: 1024
+    win_length: 1024
+    hop_length: 256
+    n_mels: 80
+    fmin: 0
+    fmax: "None"
+
+model_params:
+  dit_type: "DiT" # uDiT or DiT
+  reg_loss_type: "l1" # l1 or l2
+
+  timbre_shifter:
+    se_db_path: "./modules/openvoice/checkpoints_v2/converter/se_db.pt"
+    ckpt_path: './modules/openvoice/checkpoints_v2/converter'
+
+  speech_tokenizer:
+    type: 'whisper'
+    name: "openai/whisper-small"
+
+  style_encoder:
+    dim: 192
+    campplus_path: "campplus_cn_common.bin"
+
+  vocoder:
+    type: "bigvgan"
+    name: "nvidia/bigvgan_v2_22khz_80band_256x"
+
+  length_regulator:
+    channels: 512
+    is_discrete: false
+    in_channels: 768
+    content_codebook_size: 2048
+    sampling_ratios: [1, 1, 1, 1]
+    vector_quantize: false
+    n_codebooks: 1
+    quantizer_dropout: 0.0
+    f0_condition: false
+    n_f0_bins: 512
+
+  DiT:
+    hidden_dim: 512
+    num_heads: 8
+    depth: 13
+    class_dropout_prob: 0.1
+    block_size: 8192
+    in_channels: 80
+    style_condition: true
+    final_layer_type: 'wavenet'
+    target: 'mel' # mel or codec
+    content_dim: 512
+    content_codebook_size: 1024
+    content_type: 'discrete'
+    f0_condition: false
+    n_f0_bins: 512
+    content_codebooks: 1
+    is_causal: false
+    long_skip_connection: true
+    zero_prompt_speech_token: false # for prompt component, do not input corresponding speech token
+    time_as_token: false
+    style_as_token: false
+    uvit_skip_connection: true
+    add_resblock_in_transformer: false
+
+  wavenet:
+    hidden_dim: 512
+    num_layers: 8
+    kernel_size: 5
+    dilation_rate: 1
+    p_dropout: 0.2
+    style_condition: true
+
+loss_params:
+  base_lr: 0.0001
+  lambda_mel: 45
+  lambda_kl: 1.0

models/Seed-VC (memescreamer)/config_dit_mel_seed_uvit_xlsr_tiny.yml

@@ -0,0 +1,82 @@
+log_dir: "runs/run_mel_seed_uvit_xlsr_tiny"
+save_freq: 1
+log_interval: 10
+save_interval: 500
+device: "cuda"
+epochs: 1000 # number of epochs for first stage training (pre-training)
+batch_size: 2
+batch_length: 100 # maximum duration of audio in a batch (in seconds)
+max_len: 80 # maximum number of frames
+pretrained_model: "DiT_uvit_tat_xlsr_ema.pth"
+pretrained_encoder: ""
+load_only_params: False # set to true if do not want to load epoch numbers and optimizer parameters
+
+preprocess_params:
+  sr: 22050
+  spect_params:
+    n_fft: 1024
+    win_length: 1024
+    hop_length: 256
+    n_mels: 80
+    fmin: 0
+    fmax: 8000
+
+model_params:
+  dit_type: "DiT" # uDiT or DiT
+  reg_loss_type: "l1" # l1 or l2
+  diffusion_type: "flow"
+
+  timbre_shifter:
+    se_db_path: "./modules/openvoice/checkpoints_v2/converter/se_db.pt"
+    ckpt_path: './modules/openvoice/checkpoints_v2/converter'
+
+  vocoder:
+    type: "hifigan"
+
+  speech_tokenizer:
+    type: 'xlsr'
+    output_layer: 12
+    name: 'facebook/wav2vec2-xls-r-300m'
+
+  style_encoder:
+    dim: 192
+    campplus_path: "campplus_cn_common.bin"
+
+  length_regulator:
+    channels: 384
+    is_discrete: false
+    in_channels: 1024
+    content_codebook_size: 1024
+    sampling_ratios: [1, 1, 1, 1]
+    vector_quantize: false
+    n_codebooks: 2
+    quantizer_dropout: 0.0
+    f0_condition: false
+    n_f0_bins: 512
+
+  DiT:
+    hidden_dim: 384
+    num_heads: 6
+    depth: 9
+    class_dropout_prob: 0.1
+    block_size: 8192
+    in_channels: 80
+    style_condition: true
+    final_layer_type: 'mlp'
+    target: 'mel' # mel or betavae
+    content_dim: 384
+    content_codebook_size: 1024
+    content_type: 'discrete'
+    f0_condition: false
+    n_f0_bins: 512
+    content_codebooks: 1
+    is_causal: false
+    long_skip_connection: false
+    zero_prompt_speech_token: false # for prompt component, do not input corresponding speech token
+    time_as_token: true
+    style_as_token: true
+    uvit_skip_connection: true
+    add_resblock_in_transformer: false
+
+loss_params:
+  base_lr: 0.0001

models/Seed-VC (memescreamer)/config_dit_mel_seed_wavenet.yml

@@ -0,0 +1,79 @@
+log_dir: "./runs/run_dit_mel_seed"
+save_freq: 1
+log_interval: 10
+save_interval: 1000
+device: "cuda"
+epochs: 1000 # number of epochs for first stage training (pre-training)
+batch_size: 4
+batch_length: 100 # maximum duration of audio in a batch (in seconds)
+max_len: 80 # maximum number of frames
+pretrained_model: ""
+pretrained_encoder: ""
+load_only_params: False # set to true if do not want to load epoch numbers and optimizer parameters
+
+F0_path: "modules/JDC/bst.t7"
+
+preprocess_params:
+  sr: 22050
+  spect_params:
+    n_fft: 1024
+    win_length: 1024
+    hop_length: 256
+    n_mels: 80
+
+model_params:
+  dit_type: "DiT" # uDiT or DiT
+  reg_loss_type: "l2" # l1 or l2
+
+  speech_tokenizer:
+    path: "speech_tokenizer_v1.onnx"
+
+  style_encoder:
+    dim: 192
+    campplus_path: "campplus_cn_common.bin"
+
+  DAC:
+    encoder_dim: 64
+    encoder_rates: [2, 5, 5, 6]
+    decoder_dim: 1536
+    decoder_rates: [ 6, 5, 5, 2 ]
+    sr: 24000
+
+  length_regulator:
+    channels: 768
+    is_discrete: true
+    content_codebook_size: 4096
+    in_frame_rate: 50
+    out_frame_rate: 80
+    sampling_ratios: [1, 1, 1, 1]
+
+  DiT:
+    hidden_dim: 768
+    num_heads: 12
+    depth: 12
+    class_dropout_prob: 0.1
+    block_size: 4096
+    in_channels: 80
+    style_condition: true
+    final_layer_type: 'wavenet'
+    target: 'mel' # mel or codec
+    content_dim: 768
+    content_codebook_size: 1024
+    content_type: 'discrete'
+    f0_condition: false
+    n_f0_bins: 512
+    content_codebooks: 1
+    is_causal: false
+    long_skip_connection: true
+    zero_prompt_speech_token: false # for prompt component, do not input corresponding speech token
+
+  wavenet:
+    hidden_dim: 768
+    num_layers: 8
+    kernel_size: 5
+    dilation_rate: 1
+    p_dropout: 0.2
+    style_condition: true
+
+loss_params:
+  base_lr: 0.0001

models/Seed-VC (memescreamer)/hifigan.yml

@@ -0,0 +1,25 @@
+hift:
+    in_channels: 80
+    base_channels: 512
+    nb_harmonics: 8
+    sampling_rate: 22050
+    nsf_alpha: 0.1
+    nsf_sigma: 0.003
+    nsf_voiced_threshold: 10
+    upsample_rates: [8, 8]
+    upsample_kernel_sizes: [16, 16]
+    istft_params:
+        n_fft: 16
+        hop_len: 4
+    resblock_kernel_sizes: [3, 7, 11]
+    resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5], [1, 3, 5]]
+    source_resblock_kernel_sizes: [7, 11]
+    source_resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5]]
+    lrelu_slope: 0.1
+    audio_limit: 0.99
+f0_predictor:
+    num_class: 1
+    in_channels: 80
+    cond_channels: 512
+
+pretrained_model_path: "hift.pt"

models/Seed-VC (memescreamer)/hift.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91e679b6ca1eff71187ffb4f3ab0444935594cdcc20a9bd12afad111ef8d6012
+size 81896716

models/Seed-VC (memescreamer)/rmvpe/rmvpe.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d62215f4306e3ca278246188607209f09af3dc77ed4232efdd069798c4ec193
+size 181184272

models/Seed-VC (memescreamer)/se_db.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a76e1ad4fac609cfa1a5198bf789427c5e7822d656fab1ffed6cd80fc0d381b7
+size 102405286

models/Seed-VC (memescreamer)/source.txt

@@ -0,0 +1 @@
+https://huggingface.co/memescreamer/Seed-VC

models/Seed-VC (memescreamer)/v2/ar_base.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42e2afabf97e5d856580af1f81a9705423b4fb21bb9475cece9925b62355cc5c
+size 358846893

models/Seed-VC (memescreamer)/v2/cfm_small.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73ae18615dc8cb01710086629973198226dedc63ebeaab0aee17863d767b3474
+size 353349694

models/malayalam-seedvc-v2/.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

models/malayalam-seedvc-v2/AR_epoch_00666_step_50000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a7e720aebeb993d4562a05857a547db24a1062fbd8260f06819fa317f4f1c05
+size 1556030166

models/malayalam-seedvc-v2/CFM_epoch_00666_step_50000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a3d743335e5cb034ea742743456712fc3936e97c04b7475a2d5311cd12f71e6
+size 1574488384

models/malayalam-seedvc-v2/languages.txt

@@ -0,0 +1 @@
+Malayalam

models/malayalam-seedvc-v2/source.txt

@@ -0,0 +1 @@
+https://huggingface.co/Bajiyo/malayalam-seedvc-v2

models/malayalam-seedvc-v2/vc_wrapper.yaml

@@ -0,0 +1,105 @@
+_target_: modules.v2.vc_wrapper.VoiceConversionWrapper
+sr: 22050
+hop_size: 256
+mel_fn:
+  _target_: modules.audio.mel_spectrogram
+  _partial_: true
+  n_fft: 1024
+  win_size: 1024
+  hop_size: 256
+  num_mels: 80
+  sampling_rate: 22050
+  fmin: 0
+  fmax: null
+  center: False
+cfm:
+  _target_: modules.v2.cfm.CFM
+  estimator:
+    _target_: modules.v2.dit_wrapper.DiT
+    time_as_token: true
+    style_as_token: true
+    uvit_skip_connection: false
+    block_size: 8192
+    depth: 13
+    num_heads: 8
+    hidden_dim: 512
+    in_channels: 80
+    content_dim: 512
+    style_encoder_dim: 192
+    class_dropout_prob: 0.1
+    dropout_rate: 0.0
+    attn_dropout_rate: 0.0
+cfm_length_regulator:
+  _target_: modules.v2.length_regulator.InterpolateRegulator
+  channels: 512
+  is_discrete: true
+  codebook_size: 2048
+  sampling_ratios: [ 1, 1, 1, 1 ]
+  f0_condition: false
+ar:
+  _target_: modules.v2.ar.NaiveWrapper
+  model:
+    _target_: modules.v2.ar.NaiveTransformer
+    config:
+      _target_: modules.v2.ar.NaiveModelArgs
+      dropout: 0.0
+      rope_base: 10000.0
+      dim: 768
+      head_dim: 64
+      n_local_heads: 2
+      intermediate_size: 2304
+      n_head: 12
+      n_layer: 12
+      vocab_size: 2049  # 1 + 1 for eos
+ar_length_regulator:
+  _target_: modules.v2.length_regulator.InterpolateRegulator
+  channels: 768
+  is_discrete: true
+  codebook_size: 32
+  sampling_ratios: [ ]
+  f0_condition: false
+style_encoder:
+  _target_: modules.campplus.DTDNN.CAMPPlus
+  feat_dim: 80
+  embedding_size: 192
+content_extractor_narrow:
+  _target_: modules.astral_quantization.default_model.AstralQuantizer
+  tokenizer_name: "openai/whisper-small"
+  ssl_model_name: "facebook/hubert-large-ll60k"
+  ssl_output_layer: 18
+  skip_ssl: true
+  encoder: &bottleneck_encoder
+    _target_: modules.astral_quantization.convnext.ConvNeXtV2Stage
+    dim: 512
+    num_blocks: 12
+    intermediate_dim: 1536
+    dilation: 1
+    input_dim: 1024
+  quantizer:
+    _target_: modules.astral_quantization.bsq.BinarySphericalQuantize
+    codebook_size: 32  # codebook size, must be a power of 2
+    dim: 512
+    entropy_loss_weight: 0.1
+    diversity_gamma: 1.0
+    spherical: True
+    enable_entropy_loss: True
+    soft_entropy_loss: True
+content_extractor_wide:
+  _target_: modules.astral_quantization.default_model.AstralQuantizer
+  tokenizer_name: "openai/whisper-small"
+  ssl_model_name: "facebook/hubert-large-ll60k"
+  ssl_output_layer: 18
+  encoder: *bottleneck_encoder
+  quantizer:
+    _target_: modules.astral_quantization.bsq.BinarySphericalQuantize
+    codebook_size: 2048  # codebook size, must be a power of 2
+    dim: 512
+    entropy_loss_weight: 0.1
+    diversity_gamma: 1.0
+    spherical: True
+    enable_entropy_loss: True
+    soft_entropy_loss: True
+vocoder:
+  _target_: modules.bigvgan.bigvgan.BigVGAN.from_pretrained
+  pretrained_model_name_or_path: "nvidia/bigvgan_v2_22khz_80band_256x"
+  use_cuda_kernel: false

models/polyphony-seed-vc/.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

models/polyphony-seed-vc/.hydra/config.yaml

@@ -0,0 +1,100 @@
+pretrained_model: DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema.pth
+log_interval: 10
+preprocess_params:
+  sr: 44100
+  spect_params:
+    n_fft: 2048
+    win_length: 2048
+    hop_length: 512
+    n_mels: 128
+    fmin: 0
+    fmax: None
+model_params:
+  dit_type: DiT
+  reg_loss_type: l1
+  timbre_shifter:
+    se_db_path: ./modules/openvoice/checkpoints_v2/converter/se_db.pt
+    ckpt_path: ./modules/openvoice/checkpoints_v2/converter
+  vocoder:
+    type: bigvgan
+    name: nvidia/bigvgan_v2_44khz_128band_512x
+  speech_tokenizer:
+    type: whisper
+    name: openai/whisper-small
+  style_encoder:
+    dim: 192
+    campplus_path: campplus_cn_common.bin
+  DAC:
+    encoder_dim: 64
+    encoder_rates:
+    - 2
+    - 5
+    - 5
+    - 6
+    decoder_dim: 1536
+    decoder_rates:
+    - 6
+    - 5
+    - 5
+    - 2
+    sr: 24000
+  length_regulator:
+    channels: 768
+    is_discrete: false
+    in_channels: 768
+    content_codebook_size: 2048
+    sampling_ratios:
+    - 1
+    - 1
+    - 1
+    - 1
+    vector_quantize: false
+    n_codebooks: 1
+    quantizer_dropout: 0.0
+    f0_condition: true
+    n_f0_bins: 256
+  DiT:
+    hidden_dim: 768
+    num_heads: 12
+    depth: 17
+    class_dropout_prob: 0.1
+    block_size: 8192
+    in_channels: 128
+    style_condition: true
+    final_layer_type: mlp
+    target: mel
+    content_dim: 768
+    content_codebook_size: 1024
+    content_type: discrete
+    f0_condition: true
+    n_f0_bins: 256
+    content_codebooks: 1
+    is_causal: false
+    long_skip_connection: false
+    zero_prompt_speech_token: false
+    time_as_token: false
+    style_as_token: false
+    uvit_skip_connection: true
+    add_resblock_in_transformer: false
+  wavenet:
+    hidden_dim: 768
+    num_layers: 8
+    kernel_size: 5
+    dilation_rate: 1
+    p_dropout: 0.2
+    style_condition: true
+loss_params:
+  base_lr: 0.0001
+  lambda_mel: 45
+  lambda_kl: 1.0
+run_name: my_run-hydra-nfdsnflksdjflkds
+batch_size: 2
+max_steps: 10000
+max_epochs: 1000
+save_every: 500
+eval_every: 1
+num_workers: 0
+require_features: true
+pretrained_ckpt: null
+tensorboard_logging: true
+mlflow_logging: false

models/polyphony-seed-vc/.hydra/hydra.yaml

@@ -0,0 +1,155 @@
+hydra:
+  run:
+    dir: runs/${run_name}_${now:%Y-%m-%d_%H-%M-%S}
+  sweep:
+    dir: multirun/${now:%Y-%m-%d}/${now:%H-%M-%S}
+    subdir: ${hydra.job.num}
+  launcher:
+    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
+  sweeper:
+    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
+    max_batch_size: null
+    params: null
+  help:
+    app_name: ${hydra.job.name}
+    header: '${hydra.help.app_name} is powered by Hydra.
+
+      '
+    footer: 'Powered by Hydra (https://hydra.cc)
+
+      Use --hydra-help to view Hydra specific help
+
+      '
+    template: '${hydra.help.header}
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (group=option)
+
+
+      $APP_CONFIG_GROUPS
+
+
+      == Config ==
+
+      Override anything in the config (foo.bar=value)
+
+
+      $CONFIG
+
+
+      ${hydra.help.footer}
+
+      '
+  hydra_help:
+    template: 'Hydra (${hydra.runtime.version})
+
+      See https://hydra.cc for more info.
+
+
+      == Flags ==
+
+      $FLAGS_HELP
+
+
+      == Configuration groups ==
+
+      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
+      to command line)
+
+
+      $HYDRA_CONFIG_GROUPS
+
+
+      Use ''--cfg hydra'' to Show the Hydra config.
+
+      '
+    hydra_help: ???
+  hydra_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][HYDRA] %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+    root:
+      level: INFO
+      handlers:
+      - console
+    loggers:
+      logging_example:
+        level: DEBUG
+    disable_existing_loggers: false
+  job_logging:
+    version: 1
+    formatters:
+      simple:
+        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
+    handlers:
+      console:
+        class: logging.StreamHandler
+        formatter: simple
+        stream: ext://sys.stdout
+      file:
+        class: logging.FileHandler
+        formatter: simple
+        filename: ${hydra.runtime.output_dir}/${hydra.job.name}.log
+    root:
+      level: INFO
+      handlers:
+      - console
+      - file
+    disable_existing_loggers: false
+  env: {}
+  mode: RUN
+  searchpath: []
+  callbacks: {}
+  output_subdir: .hydra
+  overrides:
+    hydra:
+    - hydra.mode=RUN
+    task: []
+  job:
+    name: train
+    chdir: false
+    override_dirname: ''
+    id: ???
+    num: ???
+    config_name: config
+    env_set: {}
+    env_copy: []
+    config:
+      override_dirname:
+        kv_sep: '='
+        item_sep: ','
+        exclude_keys: []
+  runtime:
+    version: 1.3.2
+    version_base: '1.3'
+    cwd: /workspace/polyphony-seed-vc
+    config_sources:
+    - path: hydra.conf
+      schema: pkg
+      provider: hydra
+    - path: /workspace/polyphony-seed-vc/configs
+      schema: file
+      provider: main
+    - path: ''
+      schema: structured
+      provider: schema
+    output_dir: /workspace/polyphony-seed-vc/runs/my_run-hydra-nfdsnflksdjflkds_2026-03-31_09-33-42
+    choices:
+      model: uvit_whisper_44k
+      hydra/env: default
+      hydra/callbacks: null
+      hydra/job_logging: default
+      hydra/hydra_logging: default
+      hydra/hydra_help: default
+      hydra/help: default
+      hydra/sweeper: basic
+      hydra/launcher: basic
+      hydra/output: default
+  verbose: false

models/polyphony-seed-vc/.hydra/overrides.yaml

@@ -0,0 +1 @@
+[]

models/polyphony-seed-vc/DiT_epoch_00052_step_09000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c45007f399556355beb653ebc46952947c71cfef26ff5f1402e2dcf1dcd916a
+size 2246796075

models/polyphony-seed-vc/DiT_epoch_00055_step_09500.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8cbccbbb94c152b16394fd2ef52f65e2179629734eb1341cd7fa4da5e30f6a98
+size 2246796075

models/polyphony-seed-vc/events.out.tfevents.1774949625.85bdaf5299a2.4433.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5a20464a3e2876dec659afbed2698613464cd75fd12efd9831c375eee92ac68
+size 51905

models/polyphony-seed-vc/ft_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c03d59f289a60e95405f0a3ba839882fc60f7c61b323d77ade48d0bdc8cb0d1
+size 785751672

models/polyphony-seed-vc/source.txt

@@ -0,0 +1 @@
+https://huggingface.co/MaksymHalych/polyphony-seed-vc

models/polyphony-seed-vc/train.log

@@ -0,0 +1,81 @@
+[2026-03-31 09:33:45,418][__main__][INFO] - Loaded HF checkpoint from ./checkpoints/models--Plachta--Seed-VC/snapshots/257283f9f41585055e8f858fba4fd044e5caed6e/DiT_seed_v2_uvit_whisper_base_f0_44k_bigvgan_pruned_ft_ema.pth
+[2026-03-31 09:34:45,467][__main__][INFO] - Eval loss at epoch 1: 0.5503
+[2026-03-31 09:35:43,500][__main__][INFO] - Eval loss at epoch 2: 0.5421
+[2026-03-31 09:36:37,482][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:36:44,539][__main__][INFO] - Eval loss at epoch 3: 0.5184
+[2026-03-31 09:37:42,581][__main__][INFO] - Eval loss at epoch 4: 0.5492
+[2026-03-31 09:38:41,957][__main__][INFO] - Eval loss at epoch 5: 0.5689
+[2026-03-31 09:39:32,846][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:39:43,004][__main__][INFO] - Eval loss at epoch 6: 0.5222
+[2026-03-31 09:40:41,991][__main__][INFO] - Eval loss at epoch 7: 0.5251
+[2026-03-31 09:41:41,652][__main__][INFO] - Eval loss at epoch 8: 0.5492
+[2026-03-31 09:42:26,318][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:42:41,890][__main__][INFO] - Eval loss at epoch 9: 0.5331
+[2026-03-31 09:43:41,666][__main__][INFO] - Eval loss at epoch 10: 0.5145
+[2026-03-31 09:44:40,606][__main__][INFO] - Eval loss at epoch 11: 0.5239
+[2026-03-31 09:45:20,529][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:45:39,905][__main__][INFO] - Eval loss at epoch 12: 0.5110
+[2026-03-31 09:46:39,583][__main__][INFO] - Eval loss at epoch 13: 0.5125
+[2026-03-31 09:47:39,332][__main__][INFO] - Eval loss at epoch 14: 0.5054
+[2026-03-31 09:48:18,655][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:48:39,514][__main__][INFO] - Eval loss at epoch 15: 0.5371
+[2026-03-31 09:49:37,754][__main__][INFO] - Eval loss at epoch 16: 0.4939
+[2026-03-31 09:50:36,564][__main__][INFO] - Eval loss at epoch 17: 0.5318
+[2026-03-31 09:51:12,058][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:51:35,973][__main__][INFO] - Eval loss at epoch 18: 0.5058
+[2026-03-31 09:52:34,475][__main__][INFO] - Eval loss at epoch 19: 0.5134
+[2026-03-31 09:53:34,121][__main__][INFO] - Eval loss at epoch 20: 0.5317
+[2026-03-31 09:54:08,935][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:54:34,447][__main__][INFO] - Eval loss at epoch 21: 0.5369
+[2026-03-31 09:55:33,298][__main__][INFO] - Eval loss at epoch 22: 0.5119
+[2026-03-31 09:56:32,735][__main__][INFO] - Eval loss at epoch 23: 0.5454
+[2026-03-31 09:57:03,544][__main__][INFO] - Saving checkpoint
+[2026-03-31 09:57:33,559][__main__][INFO] - Eval loss at epoch 24: 0.5248
+[2026-03-31 09:58:33,395][__main__][INFO] - Eval loss at epoch 25: 0.5219
+[2026-03-31 09:59:32,889][__main__][INFO] - Eval loss at epoch 26: 0.5220
+[2026-03-31 09:59:59,173][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:00:32,473][__main__][INFO] - Eval loss at epoch 27: 0.5058
+[2026-03-31 10:01:31,127][__main__][INFO] - Eval loss at epoch 28: 0.5103
+[2026-03-31 10:02:29,941][__main__][INFO] - Eval loss at epoch 29: 0.5056
+[2026-03-31 10:02:54,663][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:03:30,831][__main__][INFO] - Eval loss at epoch 30: 0.5360
+[2026-03-31 10:04:29,601][__main__][INFO] - Eval loss at epoch 31: 0.4924
+[2026-03-31 10:05:29,342][__main__][INFO] - Eval loss at epoch 32: 0.4983
+[2026-03-31 10:05:48,616][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:06:30,850][__main__][INFO] - Eval loss at epoch 33: 0.4768
+[2026-03-31 10:07:30,413][__main__][INFO] - Eval loss at epoch 34: 0.4945
+[2026-03-31 10:08:29,116][__main__][INFO] - Eval loss at epoch 35: 0.4879
+[2026-03-31 10:08:46,756][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:09:29,204][__main__][INFO] - Eval loss at epoch 36: 0.5048
+[2026-03-31 10:10:28,796][__main__][INFO] - Eval loss at epoch 37: 0.5370
+[2026-03-31 10:11:28,681][__main__][INFO] - Eval loss at epoch 38: 0.4983
+[2026-03-31 10:11:41,783][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:12:29,540][__main__][INFO] - Eval loss at epoch 39: 0.4753
+[2026-03-31 10:13:29,204][__main__][INFO] - Eval loss at epoch 40: 0.5071
+[2026-03-31 10:14:27,339][__main__][INFO] - Eval loss at epoch 41: 0.4829
+[2026-03-31 10:14:38,158][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:15:27,468][__main__][INFO] - Eval loss at epoch 42: 0.4967
+[2026-03-31 10:16:26,999][__main__][INFO] - Eval loss at epoch 43: 0.4897
+[2026-03-31 10:17:25,940][__main__][INFO] - Eval loss at epoch 44: 0.4786
+[2026-03-31 10:17:32,914][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:18:25,264][__main__][INFO] - Eval loss at epoch 45: 0.4875
+[2026-03-31 10:19:23,252][__main__][INFO] - Eval loss at epoch 46: 0.4990
+[2026-03-31 10:20:23,505][__main__][INFO] - Eval loss at epoch 47: 0.5173
+[2026-03-31 10:20:26,632][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:21:21,067][__main__][INFO] - Eval loss at epoch 48: 0.4705
+[2026-03-31 10:22:20,641][__main__][INFO] - Eval loss at epoch 49: 0.4964
+[2026-03-31 10:23:16,196][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:23:20,179][__main__][INFO] - Eval loss at epoch 50: 0.4836
+[2026-03-31 10:24:19,462][__main__][INFO] - Eval loss at epoch 51: 0.5069
+[2026-03-31 10:25:18,946][__main__][INFO] - Eval loss at epoch 52: 0.4988
+[2026-03-31 10:26:12,043][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:26:19,038][__main__][INFO] - Eval loss at epoch 53: 0.4600
+[2026-03-31 10:27:17,638][__main__][INFO] - Eval loss at epoch 54: 0.4965
+[2026-03-31 10:28:17,309][__main__][INFO] - Eval loss at epoch 55: 0.4579
+[2026-03-31 10:29:05,862][__main__][INFO] - Saving checkpoint
+[2026-03-31 10:29:16,698][__main__][INFO] - Eval loss at epoch 56: 0.5023
+[2026-03-31 10:30:15,627][__main__][INFO] - Eval loss at epoch 57: 0.5086
+[2026-03-31 10:31:13,689][__main__][INFO] - Eval loss at epoch 58: 0.5095
+[2026-03-31 10:32:03,283][__main__][INFO] - Eval loss at epoch 59: 0.4990
+[2026-03-31 10:32:03,283][__main__][INFO] - Saving final model
+[2026-03-31 10:32:03,744][__main__][INFO] - Final model saved at /workspace/polyphony-seed-vc/runs/my_run-hydra-nfdsnflksdjflkds_2026-03-31_09-33-42/ft_model.pth

models/seed-vc-heb/languages.txt

@@ -0,0 +1 @@
+Hebrew