Source: https://seerepo.com/r/SWivid/F5-TTS/built
Markdown source: https://seerepo.com/r/SWivid/F5-TTS/built.md
Description: F5-TTS is a Python repo. It uses AI through transformers-stream-generator, torch, torchaudio, datasets, safetensors, and 6 more. It has 13,839 written lines.

# What is F5-TTS built with?

F5-TTS is a Python repo. It uses AI through transformers-stream-generator, torch, torchaudio, datasets, safetensors, and 6 more and makes its content with torchcodec. It has 13,839 written lines.

Read on November 8, 2026, at commit 600e5fd.

Repository: https://github.com/SWivid/F5-TTS

## Frontend

### Interface

- **[Gradio](https://seerepo.com/uses/gradio)**: 2 files
- **[matplotlib](https://seerepo.com/uses/matplotlib)**: 1 file

## Backend

### AI

- **transformers-stream-generator**
- **[torch](https://seerepo.com/uses/torch)**: 18 files
- **[torchaudio](https://seerepo.com/uses/torchaudio)**: 12 files
- **[datasets](https://seerepo.com/uses/datasets)**: 10 files
- **[safetensors](https://seerepo.com/uses/safetensors)**: 4 files
- **x-transformers**: 4 files
- **[librosa](https://seerepo.com/uses/librosa)**: 2 files
- **[transformers](https://seerepo.com/uses/transformers)**: 2 files
- **vocos**: 2 files
- **[accelerate](https://seerepo.com/uses/accelerate)**: 1 file
- **[bitsandbytes](https://seerepo.com/uses/bitsandbytes)**: 1 file

## Support

### Content

- **torchcodec**

### Tools

- **click**: 2 files

### Other

- **hydra-core**
- **tqdm**: 11 files
- **NumPy**: 10 files
- **soundfile**: 8 files
- **pypinyin**: 2 files
- **pydub**: 1 file
- **tomli**: 1 file
- **unidecode**: 1 file
- **wandb**: 1 file

## Other repos with a similar stack

- [mudler/LocalAI](https://seerepo.com/r/mudler/LocalAI/built): Gradio, matplotlib, torch, torchaudio, datasets, safetensors, librosa, transformers, accelerate, bitsandbytes, torchcodec · [Compare](https://seerepo.com/compare/mudler/LocalAI/vs/SWivid/F5-TTS)
- [hiyouga/LlamaFactory](https://seerepo.com/r/hiyouga/LlamaFactory/built): Gradio, matplotlib, torch, torchaudio, datasets, safetensors, transformers, accelerate, bitsandbytes, torchcodec · [Compare](https://seerepo.com/compare/hiyouga/LlamaFactory/vs/SWivid/F5-TTS)
- [ray-project/ray](https://seerepo.com/r/ray-project/ray/built): Gradio, matplotlib, torch, torchaudio, datasets, safetensors, transformers, accelerate, bitsandbytes, torchcodec · [Compare](https://seerepo.com/compare/ray-project/ray/vs/SWivid/F5-TTS)
- [bojieli/ai-agent-book](https://seerepo.com/r/bojieli/ai-agent-book/built): matplotlib, torch, torchaudio, datasets, librosa, transformers, accelerate, bitsandbytes, torchcodec · [Compare](https://seerepo.com/compare/bojieli/ai-agent-book/vs/SWivid/F5-TTS)
- [gradio-app/gradio](https://seerepo.com/r/gradio-app/gradio/built): Gradio, matplotlib, torch, torchaudio, datasets, safetensors, librosa, transformers, accelerate · [Compare](https://seerepo.com/compare/gradio-app/gradio/vs/SWivid/F5-TTS)
- [hpcaitech/ColossalAI](https://seerepo.com/r/hpcaitech/ColossalAI/built): Gradio, matplotlib, torch, torchaudio, datasets, safetensors, transformers, accelerate, bitsandbytes · [Compare](https://seerepo.com/compare/hpcaitech/ColossalAI/vs/SWivid/F5-TTS)

## Frequently asked questions

### What language is F5-TTS written in?

F5-TTS is written in Python, with 13,839 written lines.

### Does F5-TTS use AI?

Yes. F5-TTS uses AI through transformers-stream-generator, torch, torchaudio, datasets, safetensors, x-transformers, librosa, transformers, vocos, accelerate, and bitsandbytes.

### Which repos have a stack like F5-TTS's?

The ones that share most of its stack: mudler/LocalAI, hiyouga/LlamaFactory, and ray-project/ray.

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.