---
title: KittenTTS WebGPU
emoji: 🐱
colorFrom: purple
colorTo: indigo
sdk: static
pinned: false
license: apache-2.0
---

# KittenTTS WebGPU

Text-to-speech running entirely in your browser via WebGPU/WASM. No server needed.

## Credits

- **Models**: [KittenML](https://huggingface.co/KittenML) — ultra-lightweight TTS models based on StyleTTS 2
- **Original Demo**: [KittenTTS-Demo](https://huggingface.co/spaces/KittenML/KittenTTS-Demo)
- **Transformers.js v4**: [huggingface/transformers.js](https://github.com/huggingface/transformers.js) — ML inference in the browser
- **phonemizer.js**: [xenova/phonemizer.js](https://github.com/xenova/phonemizer.js) — eSpeak-NG phonemization for the web by [Xenova](https://github.com/xenova)
- **ONNX Runtime Web**: [onnxruntime](https://onnxruntime.ai) — cross-platform ML inference
- **Kokoro Web**: [xenova/kokoro-web](https://github.com/xenova/kokoro-web) — reference implementation for browser TTS

## Development

```bash
npm install
npm run dev
```

## Build

```bash
npm run build
```

Output goes to `dist/` — deploy as a static HuggingFace Space or any static host.