Apppublic

shreyask/KittenTTS-WebGPU

sourceHugging Faceapache-2.0updated 6mo agoView on Hugging Face

2likes

App README

KittenTTS WebGPU

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

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

bash

npm install
npm run dev

bash

npm run build

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