Kani TTS (ar, de, en, es, fr, hi, ja, ko, zh)
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +25 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/.gitattributes +47 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/README.md +77 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.IQ4_XS.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q2_K.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_L.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_M.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_S.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q4_K_M.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q4_K_S.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q5_K_M.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q5_K_S.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q6_K.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q8_0.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.f16.gguf +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-GGUF/source.txt +1 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/.gitattributes +35 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/README.md +43 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/chat_template.jinja +37 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/config.json +57 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/generation_config.json +7 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/model.safetensors +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/model.safetensors.index.json +156 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/source.txt +1 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/special_tokens_map.json +23 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/tokenizer.json +0 -0
- ar,de,en,es,ko,zh/kani-tts-370m-MLX/tokenizer_config.json +0 -0
- ar,de,en,es,ko,zh/kani-tts-370m/.gitattributes +35 -0
- ar,de,en,es,ko,zh/kani-tts-370m/README.md +193 -0
- ar,de,en,es,ko,zh/kani-tts-370m/chat_template.jinja +37 -0
- ar,de,en,es,ko,zh/kani-tts-370m/config.json +58 -0
- ar,de,en,es,ko,zh/kani-tts-370m/generation_config.json +7 -0
- ar,de,en,es,ko,zh/kani-tts-370m/model.safetensors +3 -0
- ar,de,en,es,ko,zh/kani-tts-370m/source.txt +1 -0
- ar,de,en,es,ko,zh/kani-tts-370m/special_tokens_map.json +23 -0
- ar,de,en,es,ko,zh/kani-tts-370m/tokenizer.json +0 -0
- ar,de,en,es,ko,zh/kani-tts-370m/tokenizer_config.json +0 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/.gitattributes +47 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/README.md +78 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.IQ4_XS.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q2_K.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_L.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_M.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_S.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q4_K_M.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q4_K_S.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q5_K_M.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q5_K_S.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q6_K.gguf +3 -0
- ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q8_0.gguf +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,28 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.f16.gguf filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.f16.gguf filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 56 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 57 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 58 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 59 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
| 60 |
+
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt/trainer_state.json filter=lfs diff=lfs merge=lfs -text
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/.gitattributes
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
kani-tts-370m.IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
kani-tts-370m.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
kani-tts-370m.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
kani-tts-370m.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
kani-tts-370m.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
kani-tts-370m.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
kani-tts-370m.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
kani-tts-370m.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
kani-tts-370m.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
kani-tts-370m.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
kani-tts-370m.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
kani-tts-370m.f16.gguf filter=lfs diff=lfs merge=lfs -text
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: nineninesix/kani-tts-370m
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- de
|
| 6 |
+
- ar
|
| 7 |
+
- zh
|
| 8 |
+
- es
|
| 9 |
+
- ko
|
| 10 |
+
library_name: transformers
|
| 11 |
+
license: apache-2.0
|
| 12 |
+
mradermacher:
|
| 13 |
+
readme_rev: 1
|
| 14 |
+
quantized_by: mradermacher
|
| 15 |
+
---
|
| 16 |
+
## About
|
| 17 |
+
|
| 18 |
+
<!-- ### quantize_version: 2 -->
|
| 19 |
+
<!-- ### output_tensor_quantised: 1 -->
|
| 20 |
+
<!-- ### convert_type: hf -->
|
| 21 |
+
<!-- ### vocab_type: -->
|
| 22 |
+
<!-- ### tags: -->
|
| 23 |
+
<!-- ### quants: x-f16 Q4_K_S Q2_K Q8_0 Q6_K Q3_K_M Q3_K_S Q3_K_L Q4_K_M Q5_K_S Q5_K_M IQ4_XS -->
|
| 24 |
+
<!-- ### quants_skip: -->
|
| 25 |
+
<!-- ### skip_mmproj: -->
|
| 26 |
+
static quants of https://huggingface.co/nineninesix/kani-tts-370m
|
| 27 |
+
|
| 28 |
+
<!-- provided-files -->
|
| 29 |
+
|
| 30 |
+
***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#kani-tts-370m-GGUF).***
|
| 31 |
+
|
| 32 |
+
weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
|
| 33 |
+
## Usage
|
| 34 |
+
|
| 35 |
+
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
| 36 |
+
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
|
| 37 |
+
more details, including on how to concatenate multi-part files.
|
| 38 |
+
|
| 39 |
+
## Provided Quants
|
| 40 |
+
|
| 41 |
+
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
|
| 42 |
+
|
| 43 |
+
| Link | Type | Size/GB | Notes |
|
| 44 |
+
|:-----|:-----|--------:|:------|
|
| 45 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q2_K.gguf) | Q2_K | 0.3 | |
|
| 46 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q3_K_S.gguf) | Q3_K_S | 0.3 | |
|
| 47 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q3_K_M.gguf) | Q3_K_M | 0.3 | lower quality |
|
| 48 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q3_K_L.gguf) | Q3_K_L | 0.3 | |
|
| 49 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.IQ4_XS.gguf) | IQ4_XS | 0.3 | |
|
| 50 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q4_K_S.gguf) | Q4_K_S | 0.3 | fast, recommended |
|
| 51 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q4_K_M.gguf) | Q4_K_M | 0.3 | fast, recommended |
|
| 52 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q5_K_S.gguf) | Q5_K_S | 0.4 | |
|
| 53 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q5_K_M.gguf) | Q5_K_M | 0.4 | |
|
| 54 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q6_K.gguf) | Q6_K | 0.4 | very good quality |
|
| 55 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.Q8_0.gguf) | Q8_0 | 0.5 | fast, best quality |
|
| 56 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-370m-GGUF/resolve/main/kani-tts-370m.f16.gguf) | f16 | 0.8 | 16 bpw, overkill |
|
| 57 |
+
|
| 58 |
+
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 59 |
+
types (lower is better):
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
And here are Artefact2's thoughts on the matter:
|
| 64 |
+
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
|
| 65 |
+
|
| 66 |
+
## FAQ / Model Request
|
| 67 |
+
|
| 68 |
+
See https://huggingface.co/mradermacher/model_requests for some answers to
|
| 69 |
+
questions you might have and/or if you want some other model quantized.
|
| 70 |
+
|
| 71 |
+
## Thanks
|
| 72 |
+
|
| 73 |
+
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
|
| 74 |
+
me use its servers and providing upgrades to my workstation to enable
|
| 75 |
+
this work in my free time.
|
| 76 |
+
|
| 77 |
+
<!-- end -->
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.IQ4_XS.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:025e7a1ae0312043320c58b00fb7d0cc3d1e135378184e33ac68e8def477da18
|
| 3 |
+
size 224887744
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q2_K.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a843924ea5801614bfaea112f66939738a95a45705a8998af1594203eff7d8bc
|
| 3 |
+
size 173679552
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_L.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f3ba13173f2eff7e740502444cc4ee22b6ffee902fbd692565356a823228de0
|
| 3 |
+
size 216130496
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44e47d8ac638cbcfbca6c02b3f4cd9ebbf23d99a37ebff25ffa6d13a80ec695e
|
| 3 |
+
size 206234560
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q3_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ffa03eb9ddf981621727c9b71d260e0834297e1ca680ec5f90cfb5b8a652fa0
|
| 3 |
+
size 194233280
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q4_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca43e19af8430692dc2450c987999dbffec9e9aae120be632adf403b9f5f8cda
|
| 3 |
+
size 242394048
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q4_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53c7f472c5caedcfabad036d4ca52495a27489f455081c29c1c6c978e3248f36
|
| 3 |
+
size 233833408
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q5_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18b3cff1d1acf23f252b3503fb3a34490f33a501c0ab4f1a14e99e8c1a4395fc
|
| 3 |
+
size 273458112
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q5_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:680bc5fb48cb867b4f9c87c693b68a865ccd6a4b2f990e655557f796f41c5309
|
| 3 |
+
size 268305344
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q6_K.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25af2bbdfdfc06c486c798a625f138a80b19454d703378d8f1d0c4b9d81d83b1
|
| 3 |
+
size 306463680
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.Q8_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9e5a886e49c9898309c5bee9342ed1044398230ce26cdcdc019f5988bfa74f5
|
| 3 |
+
size 396020192
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/kani-tts-370m.f16.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca4678dc3491b4eeac33b37a761abef3028b4ad78e2fef55204e2f5ba3612898
|
| 3 |
+
size 742690592
|
ar,de,en,es,ko,zh/kani-tts-370m-GGUF/source.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
https://huggingface.co/mradermacher/kani-tts-370m-GGUF
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/.gitattributes
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/README.md
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- de
|
| 6 |
+
- ar
|
| 7 |
+
- zh
|
| 8 |
+
- es
|
| 9 |
+
- ko
|
| 10 |
+
pipeline_tag: text-to-speech
|
| 11 |
+
library_name: mlx
|
| 12 |
+
base_model: nineninesix/kani-tts-370m
|
| 13 |
+
tags:
|
| 14 |
+
- mlx
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
# nineninesix/kani-tts-370m-MLX
|
| 18 |
+
|
| 19 |
+
This model [nineninesix/kani-tts-370m-MLX](https://huggingface.co/nineninesix/kani-tts-370m-MLX) was
|
| 20 |
+
converted to MLX format from [nineninesix/kani-tts-370m](https://huggingface.co/nineninesix/kani-tts-370m)
|
| 21 |
+
using mlx-lm version **0.28.2**.
|
| 22 |
+
|
| 23 |
+
## Use with mlx
|
| 24 |
+
|
| 25 |
+
```bash
|
| 26 |
+
pip install mlx-lm
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
```python
|
| 30 |
+
from mlx_lm import load, generate
|
| 31 |
+
|
| 32 |
+
model, tokenizer = load("nineninesix/kani-tts-370m-MLX")
|
| 33 |
+
|
| 34 |
+
prompt = "hello"
|
| 35 |
+
|
| 36 |
+
if tokenizer.chat_template is not None:
|
| 37 |
+
messages = [{"role": "user", "content": prompt}]
|
| 38 |
+
prompt = tokenizer.apply_chat_template(
|
| 39 |
+
messages, add_generation_prompt=True
|
| 40 |
+
)
|
| 41 |
+
|
| 42 |
+
response = generate(model, tokenizer, prompt=prompt, verbose=True)
|
| 43 |
+
```
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/chat_template.jinja
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{{- bos_token -}}
|
| 2 |
+
{%- set system_prompt = "" -%}
|
| 3 |
+
{%- set ns = namespace(system_prompt="") -%}
|
| 4 |
+
{%- if messages[0]["role"] == "system" -%}
|
| 5 |
+
{%- set ns.system_prompt = messages[0]["content"] -%}
|
| 6 |
+
{%- set messages = messages[1:] -%}
|
| 7 |
+
{%- endif -%}
|
| 8 |
+
{%- if tools -%}
|
| 9 |
+
{%- set ns.system_prompt = ns.system_prompt + ("\n" if ns.system_prompt else "") + "List of tools: <|tool_list_start|>[" -%}
|
| 10 |
+
{%- for tool in tools -%}
|
| 11 |
+
{%- if tool is not string -%}
|
| 12 |
+
{%- set tool = tool | tojson -%}
|
| 13 |
+
{%- endif -%}
|
| 14 |
+
{%- set ns.system_prompt = ns.system_prompt + tool -%}
|
| 15 |
+
{%- if not loop.last -%}
|
| 16 |
+
{%- set ns.system_prompt = ns.system_prompt + ", " -%}
|
| 17 |
+
{%- endif -%}
|
| 18 |
+
{%- endfor -%}
|
| 19 |
+
{%- set ns.system_prompt = ns.system_prompt + "]<|tool_list_end|>" -%}
|
| 20 |
+
{%- endif -%}
|
| 21 |
+
{%- if ns.system_prompt -%}
|
| 22 |
+
{{- "<|im_start|>system\n" + ns.system_prompt + "<|im_end|>\n" -}}
|
| 23 |
+
{%- endif -%}
|
| 24 |
+
{%- for message in messages -%}
|
| 25 |
+
{{- "<|im_start|>" + message["role"] + "\n" -}}
|
| 26 |
+
{%- set content = message["content"] -%}
|
| 27 |
+
{%- if content is not string -%}
|
| 28 |
+
{%- set content = content | tojson -%}
|
| 29 |
+
{%- endif -%}
|
| 30 |
+
{%- if message["role"] == "tool" -%}
|
| 31 |
+
{%- set content = "<|tool_response_start|>" + content + "<|tool_response_end|>" -%}
|
| 32 |
+
{%- endif -%}
|
| 33 |
+
{{- content + "<|im_end|>\n" -}}
|
| 34 |
+
{%- endfor -%}
|
| 35 |
+
{%- if add_generation_prompt -%}
|
| 36 |
+
{{- "<|im_start|>assistant\n" -}}
|
| 37 |
+
{%- endif -%}
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Lfm2ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"block_auto_adjust_ff_dim": true,
|
| 6 |
+
"block_dim": 1024,
|
| 7 |
+
"block_ff_dim": 6656,
|
| 8 |
+
"block_ffn_dim_multiplier": 1.0,
|
| 9 |
+
"block_mlp_init_scale": 1.0,
|
| 10 |
+
"block_multiple_of": 256,
|
| 11 |
+
"block_norm_eps": 1e-05,
|
| 12 |
+
"block_out_init_scale": 1.0,
|
| 13 |
+
"block_use_swiglu": true,
|
| 14 |
+
"block_use_xavier_init": true,
|
| 15 |
+
"bos_token_id": 1,
|
| 16 |
+
"conv_L_cache": 3,
|
| 17 |
+
"conv_bias": false,
|
| 18 |
+
"conv_dim": 1024,
|
| 19 |
+
"conv_dim_out": 1024,
|
| 20 |
+
"conv_use_xavier_init": true,
|
| 21 |
+
"dtype": "float32",
|
| 22 |
+
"eos_token_id": 7,
|
| 23 |
+
"hidden_size": 1024,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"intermediate_size": 6656,
|
| 26 |
+
"layer_types": [
|
| 27 |
+
"conv",
|
| 28 |
+
"conv",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"conv",
|
| 31 |
+
"conv",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"conv",
|
| 34 |
+
"conv",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"conv",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"conv",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"conv",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"conv"
|
| 43 |
+
],
|
| 44 |
+
"max_position_embeddings": 128000,
|
| 45 |
+
"model_type": "lfm2",
|
| 46 |
+
"norm_eps": 1e-05,
|
| 47 |
+
"num_attention_heads": 16,
|
| 48 |
+
"num_heads": 16,
|
| 49 |
+
"num_hidden_layers": 16,
|
| 50 |
+
"num_key_value_heads": 8,
|
| 51 |
+
"pad_token_id": 0,
|
| 52 |
+
"rope_theta": 1000000.0,
|
| 53 |
+
"transformers_version": "4.56.0",
|
| 54 |
+
"use_cache": true,
|
| 55 |
+
"use_pos_enc": true,
|
| 56 |
+
"vocab_size": 80539
|
| 57 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 7,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.56.0"
|
| 7 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ac75652d273daf232f6a16bfd1d27b380887facb3b2debe57387f3bb4f051dc
|
| 3 |
+
size 1479404615
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,156 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 1479388160,
|
| 4 |
+
"total_parameters": 369847040
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"model.embed_tokens.weight": "model.safetensors",
|
| 8 |
+
"model.embedding_norm.weight": "model.safetensors",
|
| 9 |
+
"model.layers.0.conv.conv.weight": "model.safetensors",
|
| 10 |
+
"model.layers.0.conv.in_proj.weight": "model.safetensors",
|
| 11 |
+
"model.layers.0.conv.out_proj.weight": "model.safetensors",
|
| 12 |
+
"model.layers.0.feed_forward.w1.weight": "model.safetensors",
|
| 13 |
+
"model.layers.0.feed_forward.w2.weight": "model.safetensors",
|
| 14 |
+
"model.layers.0.feed_forward.w3.weight": "model.safetensors",
|
| 15 |
+
"model.layers.0.ffn_norm.weight": "model.safetensors",
|
| 16 |
+
"model.layers.0.operator_norm.weight": "model.safetensors",
|
| 17 |
+
"model.layers.1.conv.conv.weight": "model.safetensors",
|
| 18 |
+
"model.layers.1.conv.in_proj.weight": "model.safetensors",
|
| 19 |
+
"model.layers.1.conv.out_proj.weight": "model.safetensors",
|
| 20 |
+
"model.layers.1.feed_forward.w1.weight": "model.safetensors",
|
| 21 |
+
"model.layers.1.feed_forward.w2.weight": "model.safetensors",
|
| 22 |
+
"model.layers.1.feed_forward.w3.weight": "model.safetensors",
|
| 23 |
+
"model.layers.1.ffn_norm.weight": "model.safetensors",
|
| 24 |
+
"model.layers.1.operator_norm.weight": "model.safetensors",
|
| 25 |
+
"model.layers.10.feed_forward.w1.weight": "model.safetensors",
|
| 26 |
+
"model.layers.10.feed_forward.w2.weight": "model.safetensors",
|
| 27 |
+
"model.layers.10.feed_forward.w3.weight": "model.safetensors",
|
| 28 |
+
"model.layers.10.ffn_norm.weight": "model.safetensors",
|
| 29 |
+
"model.layers.10.operator_norm.weight": "model.safetensors",
|
| 30 |
+
"model.layers.10.self_attn.k_layernorm.weight": "model.safetensors",
|
| 31 |
+
"model.layers.10.self_attn.k_proj.weight": "model.safetensors",
|
| 32 |
+
"model.layers.10.self_attn.out_proj.weight": "model.safetensors",
|
| 33 |
+
"model.layers.10.self_attn.q_layernorm.weight": "model.safetensors",
|
| 34 |
+
"model.layers.10.self_attn.q_proj.weight": "model.safetensors",
|
| 35 |
+
"model.layers.10.self_attn.v_proj.weight": "model.safetensors",
|
| 36 |
+
"model.layers.11.conv.conv.weight": "model.safetensors",
|
| 37 |
+
"model.layers.11.conv.in_proj.weight": "model.safetensors",
|
| 38 |
+
"model.layers.11.conv.out_proj.weight": "model.safetensors",
|
| 39 |
+
"model.layers.11.feed_forward.w1.weight": "model.safetensors",
|
| 40 |
+
"model.layers.11.feed_forward.w2.weight": "model.safetensors",
|
| 41 |
+
"model.layers.11.feed_forward.w3.weight": "model.safetensors",
|
| 42 |
+
"model.layers.11.ffn_norm.weight": "model.safetensors",
|
| 43 |
+
"model.layers.11.operator_norm.weight": "model.safetensors",
|
| 44 |
+
"model.layers.12.feed_forward.w1.weight": "model.safetensors",
|
| 45 |
+
"model.layers.12.feed_forward.w2.weight": "model.safetensors",
|
| 46 |
+
"model.layers.12.feed_forward.w3.weight": "model.safetensors",
|
| 47 |
+
"model.layers.12.ffn_norm.weight": "model.safetensors",
|
| 48 |
+
"model.layers.12.operator_norm.weight": "model.safetensors",
|
| 49 |
+
"model.layers.12.self_attn.k_layernorm.weight": "model.safetensors",
|
| 50 |
+
"model.layers.12.self_attn.k_proj.weight": "model.safetensors",
|
| 51 |
+
"model.layers.12.self_attn.out_proj.weight": "model.safetensors",
|
| 52 |
+
"model.layers.12.self_attn.q_layernorm.weight": "model.safetensors",
|
| 53 |
+
"model.layers.12.self_attn.q_proj.weight": "model.safetensors",
|
| 54 |
+
"model.layers.12.self_attn.v_proj.weight": "model.safetensors",
|
| 55 |
+
"model.layers.13.conv.conv.weight": "model.safetensors",
|
| 56 |
+
"model.layers.13.conv.in_proj.weight": "model.safetensors",
|
| 57 |
+
"model.layers.13.conv.out_proj.weight": "model.safetensors",
|
| 58 |
+
"model.layers.13.feed_forward.w1.weight": "model.safetensors",
|
| 59 |
+
"model.layers.13.feed_forward.w2.weight": "model.safetensors",
|
| 60 |
+
"model.layers.13.feed_forward.w3.weight": "model.safetensors",
|
| 61 |
+
"model.layers.13.ffn_norm.weight": "model.safetensors",
|
| 62 |
+
"model.layers.13.operator_norm.weight": "model.safetensors",
|
| 63 |
+
"model.layers.14.feed_forward.w1.weight": "model.safetensors",
|
| 64 |
+
"model.layers.14.feed_forward.w2.weight": "model.safetensors",
|
| 65 |
+
"model.layers.14.feed_forward.w3.weight": "model.safetensors",
|
| 66 |
+
"model.layers.14.ffn_norm.weight": "model.safetensors",
|
| 67 |
+
"model.layers.14.operator_norm.weight": "model.safetensors",
|
| 68 |
+
"model.layers.14.self_attn.k_layernorm.weight": "model.safetensors",
|
| 69 |
+
"model.layers.14.self_attn.k_proj.weight": "model.safetensors",
|
| 70 |
+
"model.layers.14.self_attn.out_proj.weight": "model.safetensors",
|
| 71 |
+
"model.layers.14.self_attn.q_layernorm.weight": "model.safetensors",
|
| 72 |
+
"model.layers.14.self_attn.q_proj.weight": "model.safetensors",
|
| 73 |
+
"model.layers.14.self_attn.v_proj.weight": "model.safetensors",
|
| 74 |
+
"model.layers.15.conv.conv.weight": "model.safetensors",
|
| 75 |
+
"model.layers.15.conv.in_proj.weight": "model.safetensors",
|
| 76 |
+
"model.layers.15.conv.out_proj.weight": "model.safetensors",
|
| 77 |
+
"model.layers.15.feed_forward.w1.weight": "model.safetensors",
|
| 78 |
+
"model.layers.15.feed_forward.w2.weight": "model.safetensors",
|
| 79 |
+
"model.layers.15.feed_forward.w3.weight": "model.safetensors",
|
| 80 |
+
"model.layers.15.ffn_norm.weight": "model.safetensors",
|
| 81 |
+
"model.layers.15.operator_norm.weight": "model.safetensors",
|
| 82 |
+
"model.layers.2.feed_forward.w1.weight": "model.safetensors",
|
| 83 |
+
"model.layers.2.feed_forward.w2.weight": "model.safetensors",
|
| 84 |
+
"model.layers.2.feed_forward.w3.weight": "model.safetensors",
|
| 85 |
+
"model.layers.2.ffn_norm.weight": "model.safetensors",
|
| 86 |
+
"model.layers.2.operator_norm.weight": "model.safetensors",
|
| 87 |
+
"model.layers.2.self_attn.k_layernorm.weight": "model.safetensors",
|
| 88 |
+
"model.layers.2.self_attn.k_proj.weight": "model.safetensors",
|
| 89 |
+
"model.layers.2.self_attn.out_proj.weight": "model.safetensors",
|
| 90 |
+
"model.layers.2.self_attn.q_layernorm.weight": "model.safetensors",
|
| 91 |
+
"model.layers.2.self_attn.q_proj.weight": "model.safetensors",
|
| 92 |
+
"model.layers.2.self_attn.v_proj.weight": "model.safetensors",
|
| 93 |
+
"model.layers.3.conv.conv.weight": "model.safetensors",
|
| 94 |
+
"model.layers.3.conv.in_proj.weight": "model.safetensors",
|
| 95 |
+
"model.layers.3.conv.out_proj.weight": "model.safetensors",
|
| 96 |
+
"model.layers.3.feed_forward.w1.weight": "model.safetensors",
|
| 97 |
+
"model.layers.3.feed_forward.w2.weight": "model.safetensors",
|
| 98 |
+
"model.layers.3.feed_forward.w3.weight": "model.safetensors",
|
| 99 |
+
"model.layers.3.ffn_norm.weight": "model.safetensors",
|
| 100 |
+
"model.layers.3.operator_norm.weight": "model.safetensors",
|
| 101 |
+
"model.layers.4.conv.conv.weight": "model.safetensors",
|
| 102 |
+
"model.layers.4.conv.in_proj.weight": "model.safetensors",
|
| 103 |
+
"model.layers.4.conv.out_proj.weight": "model.safetensors",
|
| 104 |
+
"model.layers.4.feed_forward.w1.weight": "model.safetensors",
|
| 105 |
+
"model.layers.4.feed_forward.w2.weight": "model.safetensors",
|
| 106 |
+
"model.layers.4.feed_forward.w3.weight": "model.safetensors",
|
| 107 |
+
"model.layers.4.ffn_norm.weight": "model.safetensors",
|
| 108 |
+
"model.layers.4.operator_norm.weight": "model.safetensors",
|
| 109 |
+
"model.layers.5.feed_forward.w1.weight": "model.safetensors",
|
| 110 |
+
"model.layers.5.feed_forward.w2.weight": "model.safetensors",
|
| 111 |
+
"model.layers.5.feed_forward.w3.weight": "model.safetensors",
|
| 112 |
+
"model.layers.5.ffn_norm.weight": "model.safetensors",
|
| 113 |
+
"model.layers.5.operator_norm.weight": "model.safetensors",
|
| 114 |
+
"model.layers.5.self_attn.k_layernorm.weight": "model.safetensors",
|
| 115 |
+
"model.layers.5.self_attn.k_proj.weight": "model.safetensors",
|
| 116 |
+
"model.layers.5.self_attn.out_proj.weight": "model.safetensors",
|
| 117 |
+
"model.layers.5.self_attn.q_layernorm.weight": "model.safetensors",
|
| 118 |
+
"model.layers.5.self_attn.q_proj.weight": "model.safetensors",
|
| 119 |
+
"model.layers.5.self_attn.v_proj.weight": "model.safetensors",
|
| 120 |
+
"model.layers.6.conv.conv.weight": "model.safetensors",
|
| 121 |
+
"model.layers.6.conv.in_proj.weight": "model.safetensors",
|
| 122 |
+
"model.layers.6.conv.out_proj.weight": "model.safetensors",
|
| 123 |
+
"model.layers.6.feed_forward.w1.weight": "model.safetensors",
|
| 124 |
+
"model.layers.6.feed_forward.w2.weight": "model.safetensors",
|
| 125 |
+
"model.layers.6.feed_forward.w3.weight": "model.safetensors",
|
| 126 |
+
"model.layers.6.ffn_norm.weight": "model.safetensors",
|
| 127 |
+
"model.layers.6.operator_norm.weight": "model.safetensors",
|
| 128 |
+
"model.layers.7.conv.conv.weight": "model.safetensors",
|
| 129 |
+
"model.layers.7.conv.in_proj.weight": "model.safetensors",
|
| 130 |
+
"model.layers.7.conv.out_proj.weight": "model.safetensors",
|
| 131 |
+
"model.layers.7.feed_forward.w1.weight": "model.safetensors",
|
| 132 |
+
"model.layers.7.feed_forward.w2.weight": "model.safetensors",
|
| 133 |
+
"model.layers.7.feed_forward.w3.weight": "model.safetensors",
|
| 134 |
+
"model.layers.7.ffn_norm.weight": "model.safetensors",
|
| 135 |
+
"model.layers.7.operator_norm.weight": "model.safetensors",
|
| 136 |
+
"model.layers.8.feed_forward.w1.weight": "model.safetensors",
|
| 137 |
+
"model.layers.8.feed_forward.w2.weight": "model.safetensors",
|
| 138 |
+
"model.layers.8.feed_forward.w3.weight": "model.safetensors",
|
| 139 |
+
"model.layers.8.ffn_norm.weight": "model.safetensors",
|
| 140 |
+
"model.layers.8.operator_norm.weight": "model.safetensors",
|
| 141 |
+
"model.layers.8.self_attn.k_layernorm.weight": "model.safetensors",
|
| 142 |
+
"model.layers.8.self_attn.k_proj.weight": "model.safetensors",
|
| 143 |
+
"model.layers.8.self_attn.out_proj.weight": "model.safetensors",
|
| 144 |
+
"model.layers.8.self_attn.q_layernorm.weight": "model.safetensors",
|
| 145 |
+
"model.layers.8.self_attn.q_proj.weight": "model.safetensors",
|
| 146 |
+
"model.layers.8.self_attn.v_proj.weight": "model.safetensors",
|
| 147 |
+
"model.layers.9.conv.conv.weight": "model.safetensors",
|
| 148 |
+
"model.layers.9.conv.in_proj.weight": "model.safetensors",
|
| 149 |
+
"model.layers.9.conv.out_proj.weight": "model.safetensors",
|
| 150 |
+
"model.layers.9.feed_forward.w1.weight": "model.safetensors",
|
| 151 |
+
"model.layers.9.feed_forward.w2.weight": "model.safetensors",
|
| 152 |
+
"model.layers.9.feed_forward.w3.weight": "model.safetensors",
|
| 153 |
+
"model.layers.9.ffn_norm.weight": "model.safetensors",
|
| 154 |
+
"model.layers.9.operator_norm.weight": "model.safetensors"
|
| 155 |
+
}
|
| 156 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/source.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
https://huggingface.co/nineninesix/kani-tts-370m-MLX
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|startoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|im_end|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|pad|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ar,de,en,es,ko,zh/kani-tts-370m-MLX/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ar,de,en,es,ko,zh/kani-tts-370m/.gitattributes
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
ar,de,en,es,ko,zh/kani-tts-370m/README.md
ADDED
|
@@ -0,0 +1,193 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- hi
|
| 5 |
+
- en
|
| 6 |
+
- de
|
| 7 |
+
- ar
|
| 8 |
+
- zh
|
| 9 |
+
- es
|
| 10 |
+
- ko
|
| 11 |
+
pipeline_tag: text-to-speech
|
| 12 |
+
library_name: transformers
|
| 13 |
+
base_model:
|
| 14 |
+
- nineninesix/kani-tts-450m-0.2-pt
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
<p>
|
| 19 |
+
<img src="https://www.nineninesix.ai/kitty.png" alt="Logo" width="200" height="200">
|
| 20 |
+
</p>
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
# KaniTTS
|
| 24 |
+
|
| 25 |
+
[](https://opensource.org/licenses/Apache-2.0)
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
A high-speed, high-fidelity Text-to-Speech model optimized for real-time conversational AI applications.
|
| 29 |
+
|
| 30 |
+
## Overview
|
| 31 |
+
|
| 32 |
+
KaniTTS uses a two-stage pipeline combining a large language model with an efficient audio codec for exceptional speed and audio quality. The architecture generates compressed token representations through a backbone LLM, then rapidly synthesizes waveforms via neural audio codec, achieving extremely low latency.
|
| 33 |
+
|
| 34 |
+
**Key Specifications:**
|
| 35 |
+
- **Model Size:** 370M parameters
|
| 36 |
+
- **Sample Rate:** 22kHz
|
| 37 |
+
- **Languages:** English, German, Chinese, Korean, Arabic, Spanish
|
| 38 |
+
- **License:** Apache 2.0
|
| 39 |
+
|
| 40 |
+
## Performance
|
| 41 |
+
|
| 42 |
+
**Nvidia RTX 5080 Benchmarks:**
|
| 43 |
+
- **Latency:** ~1 second to generate 15 seconds of audio
|
| 44 |
+
- **Memory:** 2GB GPU VRAM
|
| 45 |
+
- **Quality Metrics:** MOS 4.3/5 (naturalness), WER <5% (accuracy)
|
| 46 |
+
|
| 47 |
+
**Pretraining:**
|
| 48 |
+
- **Dataset:** ~80k hours from LibriTTS, Common Voice, and Emilia
|
| 49 |
+
- **Hardware:** 8x H100 GPUs, 45 hours training time on [Lambda AI](https://lambda.ai/)
|
| 50 |
+
|
| 51 |
+
**Voices Datasets**
|
| 52 |
+
- [https://huggingface.co/datasets/nytopop/expresso-conversational](https://huggingface.co/datasets/nytopop/expresso-conversational)
|
| 53 |
+
- [https://huggingface.co/datasets/shb777/gemini-flash-2.0-speech](https://huggingface.co/datasets/shb777/gemini-flash-2.0-speech)
|
| 54 |
+
- [https://huggingface.co/datasets/jazza234234/david-dataset](https://huggingface.co/datasets/jazza234234/david-dataset)
|
| 55 |
+
- [https://huggingface.co/datasets/reach-vb/jenny_tts_dataset](https://huggingface.co/datasets/reach-vb/jenny_tts_dataset)
|
| 56 |
+
- [https://huggingface.co/datasets/MBZUAI/ArVoice](https://huggingface.co/datasets/MBZUAI/ArVoice)
|
| 57 |
+
- [https://huggingface.co/datasets/Thorsten-Voice/TV-44kHz-Full](https://huggingface.co/datasets/Thorsten-Voice/TV-44kHz-Full)
|
| 58 |
+
- [https://huggingface.co/datasets/SinclairSchneider/german_voice_cb](https://huggingface.co/datasets/SinclairSchneider/german_voice_cb)
|
| 59 |
+
- [https://huggingface.co/datasets/Bingsu/KSS_Dataset](https://huggingface.co/datasets/Bingsu/KSS_Dataset)
|
| 60 |
+
- [https://huggingface.co/datasets/ciempiess/ciempiess_fem](https://huggingface.co/datasets/ciempiess/ciempiess_fem)
|
| 61 |
+
- [https://huggingface.co/datasets/TingChen-ppmc/Shanghai_Dialect_TTS_openai](https://huggingface.co/datasets/TingChen-ppmc/Shanghai_Dialect_TTS_openai)
|
| 62 |
+
- [https://huggingface.co/datasets/boniromou/zh-yue-tts-dataset](https://huggingface.co/datasets/boniromou/zh-yue-tts-dataset)
|
| 63 |
+
- [https://huggingface.co/datasets/zeeshanparvez/andrew-v3](https://huggingface.co/datasets/zeeshanparvez/andrew-v3)
|
| 64 |
+
|
| 65 |
+
**Voices:**
|
| 66 |
+
- `david` — David, English (British)
|
| 67 |
+
- `puck` — Puck, English (Gemini)
|
| 68 |
+
- `kore` — Kore, English (Gemini)
|
| 69 |
+
- `andrew` — Andrew, English
|
| 70 |
+
- `jenny` — Jenny, English (Irish)
|
| 71 |
+
- `simon` — Simon, English
|
| 72 |
+
- `katie` — Katie, English
|
| 73 |
+
- `seulgi` — Seulgi, Korean
|
| 74 |
+
- `bert` — Bert, German
|
| 75 |
+
- `thorsten` — Thorsten, German (Hessisch)
|
| 76 |
+
- `maria` — Maria, Spanish
|
| 77 |
+
- `mei` — Mei, Chinese (Cantonese)
|
| 78 |
+
- `ming` — Ming, Chinese (Shanghai OpenAI)
|
| 79 |
+
- `karim` — Karim, Arabic
|
| 80 |
+
- `nur` — Nur, Arabic
|
| 81 |
+
|
| 82 |
+
## Audio Examples
|
| 83 |
+
|
| 84 |
+
| Text | Audio |
|
| 85 |
+
|---|---|
|
| 86 |
+
| I do believe Marsellus Wallace, MY husband, YOUR boss, told you to take me out and do WHATEVER I WANTED. | <audio controls><source src="https://github.com/nineninesix-ai/kani-tts/raw/refs/heads/main/public/mia.wav" type="audio/mpeg"></audio> |
|
| 87 |
+
| What do we say to the god of death? Not today! | <audio controls><source src="https://github.com/nineninesix-ai/kani-tts/raw/refs/heads/main/public/arya.wav" type="audio/wav"></audio> |
|
| 88 |
+
| What do you call a lawyer with an IQ of 60? Your honor | <audio controls><source src="https://github.com/nineninesix-ai/kani-tts/raw/refs/heads/main/public/saul.wav" type="audio/wav"></audio> |
|
| 89 |
+
| You mean, let me understand this cause, you know maybe it's me, it's a little fucked up maybe, but I'm funny how, I mean funny like I'm a clown, I amuse you? | <audio controls><source src="https://github.com/nineninesix-ai/kani-tts/raw/refs/heads/main/public/tommy.wav" type="audio/wav"></audio> |
|
| 90 |
+
|
| 91 |
+
## Use Cases
|
| 92 |
+
|
| 93 |
+
- **Conversational AI:** Real-time speech for chatbots and virtual assistants
|
| 94 |
+
- **Edge/Server Deployment:** Resource-efficient inference on affordable hardware
|
| 95 |
+
- **Accessibility:** Screen readers and language learning applications
|
| 96 |
+
- **Research:** Fine-tuning for specific voices, accents, or emotions
|
| 97 |
+
|
| 98 |
+
## Limitations
|
| 99 |
+
|
| 100 |
+
- Performance degrades with inputs exceeding 2000 tokens
|
| 101 |
+
- Limited expressivity without fine-tuning for specific emotions
|
| 102 |
+
- May inherit biases from training data in prosody or pronunciation
|
| 103 |
+
- Optimized primarily for English; other languages may require additional training
|
| 104 |
+
|
| 105 |
+
## Optimization Tips
|
| 106 |
+
|
| 107 |
+
- **Multilingual Performance:** Continually pretrain on target language datasets and fine-tune NanoCodec
|
| 108 |
+
- **Batch Processing:** Use batches of 8-16 for high-throughput scenarios
|
| 109 |
+
- **Hardware:** Optimized for NVIDIA Blackwell architecture GPUs
|
| 110 |
+
|
| 111 |
+
## Resources
|
| 112 |
+
|
| 113 |
+
**Models:**
|
| 114 |
+
- [Pretrained Model](https://huggingface.co/nineninesix/kani-tts-450m-0.2-pt)
|
| 115 |
+
- [Fine-tuned Model](https://huggingface.co/nineninesix/kani-tts-370m)
|
| 116 |
+
- [HuggingFace Space](https://huggingface.co/spaces/nineninesix/KaniTTS)
|
| 117 |
+
|
| 118 |
+
**Examples:**
|
| 119 |
+
- [Inference Example](https://colab.research.google.com/drive/1mvzGs7jtAMSUz8wvNlL5uFmgFEyAPjDh?usp=sharing)
|
| 120 |
+
- [Fine-tuning Example](https://colab.research.google.com/drive/1oDIPOSHW2kUoP3CGafvh9lM6j03Z-vE6?usp=sharing)
|
| 121 |
+
- [Example Dataset](https://huggingface.co/datasets/nineninesix/expresso-conversational-en-nano-codec-dataset)
|
| 122 |
+
- [GitHub Repository](https://github.com/nineninesix-ai/kani-tts)
|
| 123 |
+
|
| 124 |
+
**Links:**
|
| 125 |
+
- [Website](https://www.nineninesix.ai/)
|
| 126 |
+
- [Contact Form](https://airtable.com/appX2G2TpoRk4M5Bf/pagO2xbIOjiwulPcP/form)
|
| 127 |
+
|
| 128 |
+
## Acknowledgments
|
| 129 |
+
|
| 130 |
+
Built on top of [LiquidAI LFM2 350M](https://huggingface.co/LiquidAI/LFM2-350M) as the backbone and [Nvidia NanoCodec](https://huggingface.co/nvidia/nemo-nano-codec-22khz-0.6kbps-12.5fps) for audio processing.
|
| 131 |
+
|
| 132 |
+
## Responsible Use
|
| 133 |
+
|
| 134 |
+
**Prohibited activities include:**
|
| 135 |
+
- Illegal content or harmful, threatening, defamatory, or obscene material
|
| 136 |
+
- Hate speech, harassment, or incitement of violence
|
| 137 |
+
- Generating false or misleading information
|
| 138 |
+
- Impersonating individuals without consent
|
| 139 |
+
- Malicious activities such as spamming, phishing, or fraud
|
| 140 |
+
|
| 141 |
+
By using this model, you agree to comply with these restrictions and all applicable laws.
|
| 142 |
+
|
| 143 |
+
## Contact
|
| 144 |
+
Have a question, feedback, or need support? Please fill out our [contact form](https://airtable.com/appX2G2TpoRk4M5Bf/pagO2xbIOjiwulPcP/form) and we'll get back to you as soon as possible.
|
| 145 |
+
|
| 146 |
+
|
| 147 |
+
## Citation
|
| 148 |
+
```
|
| 149 |
+
@misc {sb_2025,
|
| 150 |
+
author = { SB },
|
| 151 |
+
title = { gemini-flash-2.0-speech },
|
| 152 |
+
year = 2025,
|
| 153 |
+
url = { https://huggingface.co/datasets/shb777/gemini-flash-2.0-speech },
|
| 154 |
+
doi = { 10.57967/hf/4237 },
|
| 155 |
+
publisher = { Hugging Face }
|
| 156 |
+
}
|
| 157 |
+
```
|
| 158 |
+
|
| 159 |
+
```
|
| 160 |
+
@misc{toyin2025arvoicemultispeakerdatasetarabic,
|
| 161 |
+
title={ArVoice: A Multi-Speaker Dataset for Arabic Speech Synthesis},
|
| 162 |
+
author={Hawau Olamide Toyin and Rufael Marew and Humaid Alblooshi and Samar M. Magdy and Hanan Aldarmaki},
|
| 163 |
+
year={2025},
|
| 164 |
+
eprint={2505.20506},
|
| 165 |
+
archivePrefix={arXiv},
|
| 166 |
+
primaryClass={cs.CL},
|
| 167 |
+
url={https://arxiv.org/abs/2505.20506},
|
| 168 |
+
}
|
| 169 |
+
```
|
| 170 |
+
|
| 171 |
+
```
|
| 172 |
+
@misc {thorsten_müller_2024,
|
| 173 |
+
author = { {Thorsten Müller} },
|
| 174 |
+
title = { TV-44kHz-Full (Revision ff427ec) },
|
| 175 |
+
year = 2024,
|
| 176 |
+
url = { https://huggingface.co/datasets/Thorsten-Voice/TV-44kHz-Full },
|
| 177 |
+
doi = { 10.57967/hf/3290 },
|
| 178 |
+
publisher = { Hugging Face }
|
| 179 |
+
}
|
| 180 |
+
```
|
| 181 |
+
|
| 182 |
+
```
|
| 183 |
+
@misc{carlosmenaciempiessfem2019,
|
| 184 |
+
title={CIEMPIESS FEM CORPUS: Audio and Transcripts of Female Speakers in Spanish.},
|
| 185 |
+
ldc_catalog_no={LDC2019S07},
|
| 186 |
+
DOI={https://doi.org/10.35111/xdx5-n815},
|
| 187 |
+
author={Hernandez Mena, Carlos Daniel},
|
| 188 |
+
journal={Linguistic Data Consortium, Philadelphia},
|
| 189 |
+
year={2019},
|
| 190 |
+
url={https://catalog.ldc.upenn.edu/LDC2019S07},
|
| 191 |
+
}
|
| 192 |
+
|
| 193 |
+
```
|
ar,de,en,es,ko,zh/kani-tts-370m/chat_template.jinja
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{{- bos_token -}}
|
| 2 |
+
{%- set system_prompt = "" -%}
|
| 3 |
+
{%- set ns = namespace(system_prompt="") -%}
|
| 4 |
+
{%- if messages[0]["role"] == "system" -%}
|
| 5 |
+
{%- set ns.system_prompt = messages[0]["content"] -%}
|
| 6 |
+
{%- set messages = messages[1:] -%}
|
| 7 |
+
{%- endif -%}
|
| 8 |
+
{%- if tools -%}
|
| 9 |
+
{%- set ns.system_prompt = ns.system_prompt + ("\n" if ns.system_prompt else "") + "List of tools: <|tool_list_start|>[" -%}
|
| 10 |
+
{%- for tool in tools -%}
|
| 11 |
+
{%- if tool is not string -%}
|
| 12 |
+
{%- set tool = tool | tojson -%}
|
| 13 |
+
{%- endif -%}
|
| 14 |
+
{%- set ns.system_prompt = ns.system_prompt + tool -%}
|
| 15 |
+
{%- if not loop.last -%}
|
| 16 |
+
{%- set ns.system_prompt = ns.system_prompt + ", " -%}
|
| 17 |
+
{%- endif -%}
|
| 18 |
+
{%- endfor -%}
|
| 19 |
+
{%- set ns.system_prompt = ns.system_prompt + "]<|tool_list_end|>" -%}
|
| 20 |
+
{%- endif -%}
|
| 21 |
+
{%- if ns.system_prompt -%}
|
| 22 |
+
{{- "<|im_start|>system\n" + ns.system_prompt + "<|im_end|>\n" -}}
|
| 23 |
+
{%- endif -%}
|
| 24 |
+
{%- for message in messages -%}
|
| 25 |
+
{{- "<|im_start|>" + message["role"] + "\n" -}}
|
| 26 |
+
{%- set content = message["content"] -%}
|
| 27 |
+
{%- if content is not string -%}
|
| 28 |
+
{%- set content = content | tojson -%}
|
| 29 |
+
{%- endif -%}
|
| 30 |
+
{%- if message["role"] == "tool" -%}
|
| 31 |
+
{%- set content = "<|tool_response_start|>" + content + "<|tool_response_end|>" -%}
|
| 32 |
+
{%- endif -%}
|
| 33 |
+
{{- content + "<|im_end|>\n" -}}
|
| 34 |
+
{%- endfor -%}
|
| 35 |
+
{%- if add_generation_prompt -%}
|
| 36 |
+
{{- "<|im_start|>assistant\n" -}}
|
| 37 |
+
{%- endif -%}
|
ar,de,en,es,ko,zh/kani-tts-370m/config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Lfm2ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"block_auto_adjust_ff_dim": true,
|
| 6 |
+
"block_dim": 1024,
|
| 7 |
+
"block_ff_dim": 6656,
|
| 8 |
+
"block_ffn_dim_multiplier": 1.0,
|
| 9 |
+
"block_mlp_init_scale": 1.0,
|
| 10 |
+
"block_multiple_of": 256,
|
| 11 |
+
"block_norm_eps": 1e-05,
|
| 12 |
+
"block_out_init_scale": 1.0,
|
| 13 |
+
"block_use_swiglu": true,
|
| 14 |
+
"block_use_xavier_init": true,
|
| 15 |
+
"bos_token_id": 1,
|
| 16 |
+
"conv_L_cache": 3,
|
| 17 |
+
"conv_bias": false,
|
| 18 |
+
"conv_dim": 1024,
|
| 19 |
+
"conv_dim_out": 1024,
|
| 20 |
+
"conv_use_xavier_init": true,
|
| 21 |
+
"dtype": "float32",
|
| 22 |
+
"eos_token_id": 7,
|
| 23 |
+
"hidden_size": 1024,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"intermediate_size": 6656,
|
| 26 |
+
"layer_types": [
|
| 27 |
+
"conv",
|
| 28 |
+
"conv",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"conv",
|
| 31 |
+
"conv",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"conv",
|
| 34 |
+
"conv",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"conv",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"conv",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"conv",
|
| 41 |
+
"full_attention",
|
| 42 |
+
"conv"
|
| 43 |
+
],
|
| 44 |
+
"max_position_embeddings": 128000,
|
| 45 |
+
"model_type": "lfm2",
|
| 46 |
+
"norm_eps": 1e-05,
|
| 47 |
+
"num_attention_heads": 16,
|
| 48 |
+
"num_heads": 16,
|
| 49 |
+
"num_hidden_layers": 16,
|
| 50 |
+
"num_key_value_heads": 8,
|
| 51 |
+
"pad_token_id": 0,
|
| 52 |
+
"rope_theta": 1000000.0,
|
| 53 |
+
"torch_dtype": "bfloat16",
|
| 54 |
+
"transformers_version": "4.54.0",
|
| 55 |
+
"use_cache": true,
|
| 56 |
+
"use_pos_enc": true,
|
| 57 |
+
"vocab_size": 80539
|
| 58 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 7,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.54.0"
|
| 7 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8083bb38ec193c15b59b4b138e455a7b6bf286688f8c353a9d10f050906da2c5
|
| 3 |
+
size 739710608
|
ar,de,en,es,ko,zh/kani-tts-370m/source.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
https://huggingface.co/nineninesix/kani-tts-370m
|
ar,de,en,es,ko,zh/kani-tts-370m/special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|startoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|im_end|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|pad|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
ar,de,en,es,ko,zh/kani-tts-370m/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ar,de,en,es,ko,zh/kani-tts-370m/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/.gitattributes
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
kani-tts-450m-0.2-pt.IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
kani-tts-450m-0.2-pt.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
kani-tts-450m-0.2-pt.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
kani-tts-450m-0.2-pt.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
kani-tts-450m-0.2-pt.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
kani-tts-450m-0.2-pt.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
kani-tts-450m-0.2-pt.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
kani-tts-450m-0.2-pt.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
kani-tts-450m-0.2-pt.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
kani-tts-450m-0.2-pt.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
kani-tts-450m-0.2-pt.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
kani-tts-450m-0.2-pt.f16.gguf filter=lfs diff=lfs merge=lfs -text
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: nineninesix/kani-tts-450m-0.2-pt
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- de
|
| 6 |
+
- fr
|
| 7 |
+
- zh
|
| 8 |
+
- ko
|
| 9 |
+
- ja
|
| 10 |
+
- ar
|
| 11 |
+
library_name: transformers
|
| 12 |
+
license: apache-2.0
|
| 13 |
+
mradermacher:
|
| 14 |
+
readme_rev: 1
|
| 15 |
+
quantized_by: mradermacher
|
| 16 |
+
---
|
| 17 |
+
## About
|
| 18 |
+
|
| 19 |
+
<!-- ### quantize_version: 2 -->
|
| 20 |
+
<!-- ### output_tensor_quantised: 1 -->
|
| 21 |
+
<!-- ### convert_type: hf -->
|
| 22 |
+
<!-- ### vocab_type: -->
|
| 23 |
+
<!-- ### tags: -->
|
| 24 |
+
<!-- ### quants: x-f16 Q4_K_S Q2_K Q8_0 Q6_K Q3_K_M Q3_K_S Q3_K_L Q4_K_M Q5_K_S Q5_K_M IQ4_XS -->
|
| 25 |
+
<!-- ### quants_skip: -->
|
| 26 |
+
<!-- ### skip_mmproj: -->
|
| 27 |
+
static quants of https://huggingface.co/nineninesix/kani-tts-450m-0.2-pt
|
| 28 |
+
|
| 29 |
+
<!-- provided-files -->
|
| 30 |
+
|
| 31 |
+
***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#kani-tts-450m-0.2-pt-GGUF).***
|
| 32 |
+
|
| 33 |
+
weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
|
| 34 |
+
## Usage
|
| 35 |
+
|
| 36 |
+
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
| 37 |
+
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
|
| 38 |
+
more details, including on how to concatenate multi-part files.
|
| 39 |
+
|
| 40 |
+
## Provided Quants
|
| 41 |
+
|
| 42 |
+
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
|
| 43 |
+
|
| 44 |
+
| Link | Type | Size/GB | Notes |
|
| 45 |
+
|:-----|:-----|--------:|:------|
|
| 46 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q2_K.gguf) | Q2_K | 0.3 | |
|
| 47 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q3_K_S.gguf) | Q3_K_S | 0.3 | |
|
| 48 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q3_K_M.gguf) | Q3_K_M | 0.3 | lower quality |
|
| 49 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q3_K_L.gguf) | Q3_K_L | 0.4 | |
|
| 50 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.IQ4_XS.gguf) | IQ4_XS | 0.4 | |
|
| 51 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q4_K_S.gguf) | Q4_K_S | 0.4 | fast, recommended |
|
| 52 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q4_K_M.gguf) | Q4_K_M | 0.4 | fast, recommended |
|
| 53 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q5_K_S.gguf) | Q5_K_S | 0.4 | |
|
| 54 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q5_K_M.gguf) | Q5_K_M | 0.4 | |
|
| 55 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q6_K.gguf) | Q6_K | 0.5 | very good quality |
|
| 56 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.Q8_0.gguf) | Q8_0 | 0.6 | fast, best quality |
|
| 57 |
+
| [GGUF](https://huggingface.co/mradermacher/kani-tts-450m-0.2-pt-GGUF/resolve/main/kani-tts-450m-0.2-pt.f16.gguf) | f16 | 1.0 | 16 bpw, overkill |
|
| 58 |
+
|
| 59 |
+
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 60 |
+
types (lower is better):
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
|
| 64 |
+
And here are Artefact2's thoughts on the matter:
|
| 65 |
+
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
|
| 66 |
+
|
| 67 |
+
## FAQ / Model Request
|
| 68 |
+
|
| 69 |
+
See https://huggingface.co/mradermacher/model_requests for some answers to
|
| 70 |
+
questions you might have and/or if you want some other model quantized.
|
| 71 |
+
|
| 72 |
+
## Thanks
|
| 73 |
+
|
| 74 |
+
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
|
| 75 |
+
me use its servers and providing upgrades to my workstation to enable
|
| 76 |
+
this work in my free time.
|
| 77 |
+
|
| 78 |
+
<!-- end -->
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.IQ4_XS.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8263af02fbe0ea8905da0d7073cd0327184af1abdb423f825289330558e6d456
|
| 3 |
+
size 268700800
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q2_K.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d39133597ec20e69f9aec65f6e586fedbcdf57866b0c336728ff0586b042636d
|
| 3 |
+
size 200740512
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_L.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea58d5d10c5bc0bfb4a9a1be7e89f6aa17d29af004a8849c63b41f12decbf945
|
| 3 |
+
size 251567520
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d01b69632b8a7dc36d79c0a17b54681ffc9cb480489b0cb0e93187141edfd834
|
| 3 |
+
size 241671584
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q3_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31401bb1250b6671c952d2d961be6d0e47dcc15e694236de78df8f7219cb89f9
|
| 3 |
+
size 229670304
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q4_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:068c05b4e66b740d8ccc4ae78a698f2b016360ccf7ccbdbeccab5202f718528f
|
| 3 |
+
size 288784352
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q4_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad2cb0848e037552405b756d912ec9458abc63ee87d53e5b8795e7ab8e316109
|
| 3 |
+
size 280223712
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q5_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08c7fbf086765fae19b02da324adecc1554548e175d317039132e056b3f2801e
|
| 3 |
+
size 330157408
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q5_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b847ad5b556d5a7f0d90c0c08f4b3f1e3f80420e52bc09c6ca706a61b4891f7
|
| 3 |
+
size 325004640
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q6_K.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b4713f524be3dba62eb48137c6b07051531d210f49736886589f146b3f1d7e55
|
| 3 |
+
size 374116288
|
ar,de,en,fr,ja,ko,zh/kani-tts-450m-0.2-pt-GGUF/kani-tts-450m-0.2-pt.Q8_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f08e68ccb3360dc03f096d4fc1398999bc3cf799e4b1b13ef1e6250261fa44f4
|
| 3 |
+
size 483646464
|