Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Fine Tune the Style-TTS2 Voice Model
| Date | Stars |
|---|---|
| 2026-07-31 | 267 |
| 2026-08-06 | 267 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# StyleTTS2 Fine-Tuning Guide
## If you want to train/finetune your model on a cloud instance then check out this - [StyleTTS2 Cloud Instance](https://dash.simplepod.ai/account/explore/55/ref-eleven/rrit_HeJcKTczE4o9).
- Make an account
- Use this template with the RTX Pro 6000 Blackwell GPU. It has 96gb of VRAM.
- I pre-installed all the packages/environment so you will just need to swap your wavs folder in and your train_list.txt/val_list.txt then you're set and you can begin training/finetuning.
- You could also pick the RTX 5090 but Stts2 heavily depends on VRAM. You will get better quality the higher the VRAM gets.
This repository provides a guide on how to prepare a dataset and execute fine-tuning using the StyleTTS2 process. https://github.com/yl4579/StyleTTS2
### If you still need to curate your dataset. You might want to checkout https://github.com/IIEleven11/Automatic-Audio-Dataset-Maker. At the end you'll need to convert it from .csv to STTSv2's .txt format (train_list.txt and val_list.txt) but that should be easy.
## Compatibility
The scripts are compatible with WSL2 and Linux. Windows requires additional dependencies and might not be worth the effort.
## Setup
### Environment Setup
1. Install conda and activate environment with Python 3.10:
- conda create --name dataset python==3.10
- conda activate dataset
## Install Pytorch
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 -U
## Install whisperx/phonemize and segmentation packages
1. ``` pip install git+https://github.com/m-bain/whisperx.git ```
2. ``` pip install phonemizer pydub pysrt tqdm ```
### Data Preparation
1. Change directory to where you have unpacked StyleTTSFineTune (You should see the makeDataset folder)
2. To make base directories you can run segmenter script. It will create the folders.
1. run
- ``` python srtsegmenter.py ```
3. Add WAV audio file/s to the audio directory (remove special characters, brackets, parenthesis to prevent issues)
4. **** This step isnt mandatory **** for the training process. You can run whisperx and segmentation without adding silence. If you do want to add silence then silencebuffer.py within the tools folder will go over your audio file, find the silent portions between sentences/breaks in speech, and add a specific length of silence to them. This could in theory provide a more accurate cut during the segmentation process. You MUST adjust the parameters within the script to fit your data. I left the values that worked for my dataset in the code, you can try them as defaults if you wish.
5. Run the following command to generate srt files for all files in the audio folder:
- Linux -
```
for i in ../audio/*.wav; do whisperx "$i" --model large-v3 --output_format srt --condition_on_previous_text True --max_line_width 250 --max_line_count 1 --segment_resolution sentence --align_model WAV2VEC2_ASR_LARGE_LV60K_960H; done
```
- Windows - in a powershell terminal copy and paste the following after verifying path to audio folder:
```
Get-ChildItem -Path 'C:\path\to\wav\folder' -Filter *.wav | ForEach-Object { whisperx $_.FullName --model large-v2 --align_model WAV2VEC2_ASR_LARGE_LV60K_960H }
```
This will generate a Whisperx .SRT file transcription of your audio. Place the srt file/s into the srt folder
### Segmentation and Transcription
1. Navigate to the main directory (You should see the folder makeDataset)
2. Within srtsegmenter.py are some variables to adjust. buffer_time and max_allowed_gap and the final if statement has a desired range you can adjust. You can try to use the defaults I have set, they worked for me. BUT! Theres a chance this will not work out well for your dataset. The process I went through would be to adjust buffer_time then run srtsegmenter.py. Go listen to the segments in order, if they are overlapping,Excerpt of 10,317 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:7a02796acdfd7dbb, llm:description: 'Fine Tune the Style-TTS2 Voice Model' (repository purpose)
matched fp:7a02796acdfd7dbb, llm:description: 'Fine Tune the Style-TTS2 Voice Model' (repository purpose)