Hugging Face's logo

the-cramer-project
/
akylai-tts-mini

Text-to-Speech
Kyrgyz
Model card Files Community
Simonlob commited on
Commit
99886ec
•
1 Parent(s): 8ba4f44

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -19
README.md CHANGED
@@ -1,21 +1,11 @@
1
- <div align="center">
 
 
 
 
 
2
 
3
-
4
-
5
- # AkylAI TTS
6
-
7
-
8
- [![python](https://img.shields.io/badge/-Python_3.10-blue?logo=python&logoColor=white)](https://www.python.org/downloads/release/python-3100/)
9
- [![pytorch](https://img.shields.io/badge/PyTorch_2.0+-ee4c2c?logo=pytorch&logoColor=white)](https://pytorch.org/get-started/locally/)
10
- [![lightning](https://img.shields.io/badge/-Lightning_2.0+-792ee5?logo=pytorchlightning&logoColor=white)](https://pytorchlightning.ai/)
11
- [![hydra](https://img.shields.io/badge/Config-Hydra_1.3-89b8cd)](https://hydra.cc/)
12
- [![black](https://img.shields.io/badge/Code%20Style-Black-black.svg?labelColor=gray)](https://black.readthedocs.io/en/stable/)
13
- [![isort](https://img.shields.io/badge/%20imports-isort-%231674b1?style=flat&labelColor=ef8336)](https://pycqa.github.io/isort/)
14
-
15
- <img src="https://github.com/simonlobgromov/Matcha-TTS/blob/main/photo_2024-04-07_15-59-52.png" height="400"/>
16
- </div>
17
-
18
- # AkylAI-TTS for Kyrgyz language
19
 
20
  We present to you a model trained in the Kyrgyz language, which has been trained on 13 hours of speech and 7,000 samples, complete with source code and training scripts. The architecture is based on Matcha-TTS.
21
  It`s a new approach to non-autoregressive neural TTS, that uses [conditional flow matching](https://arxiv.org/abs/2210.02747) (similar to [rectified flows](https://arxiv.org/abs/2209.03003)) to speed up ODE-based speech synthesis. Our method:
@@ -264,5 +254,4 @@ Other source code we would like to acknowledge:
264
  - [Hugging Face Diffusers](https://huggingface.co/): For their awesome diffusers library and its components
265
  - [Grad-TTS](https://github.com/huawei-noah/Speech-Backbones/tree/main/Grad-TTS): For the monotonic alignment search source code
266
  - [torchdyn](https://github.com/DiffEqML/torchdyn): Useful for trying other ODE solvers during research and development
267
- - [labml.ai](https://nn.labml.ai/transformers/rope/index.html): For the RoPE implementation
268
-
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - ky
5
+ pipeline_tag: text-to-speech
6
+ ---
7
 
8
+ # AkylAI, TTS for Kyrgyz language
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
 
10
  We present to you a model trained in the Kyrgyz language, which has been trained on 13 hours of speech and 7,000 samples, complete with source code and training scripts. The architecture is based on Matcha-TTS.
11
  It`s a new approach to non-autoregressive neural TTS, that uses [conditional flow matching](https://arxiv.org/abs/2210.02747) (similar to [rectified flows](https://arxiv.org/abs/2209.03003)) to speed up ODE-based speech synthesis. Our method:
 
254
  - [Hugging Face Diffusers](https://huggingface.co/): For their awesome diffusers library and its components
255
  - [Grad-TTS](https://github.com/huawei-noah/Speech-Backbones/tree/main/Grad-TTS): For the monotonic alignment search source code
256
  - [torchdyn](https://github.com/DiffEqML/torchdyn): Useful for trying other ODE solvers during research and development
257
+ - [labml.ai](https://nn.labml.ai/transformers/rope/index.html): For the RoPE implementation