A desktop application that runs Text-To-Speech (TTS) models locally on your device, without requiring an internet connection or sending your data to external servers.
Check out our release page for the latest version of the application:
This application allows you to run various Text-To-Speech models directly on your computer. It provides a simple interface to convert text to natural-sounding speech using state-of-the-art AI models while maintaining your privacy by processing everything locally.
This repository contains only the UI part of the application. To update the container running the actual models, please refer to the VoiceStudio Models Library.
- Run the VoiceStudio Backend repository at localhost:8000:
-
Install dependencies:
npm i -
Start the application:
npm start -
To build the application:
npm run make
| Model | Windows | Ubuntu/Debian | MacOS | Link |
|---|---|---|---|---|
| hexgrad/Kokoro-82M | ✅ | ✅ | ❌ | Hugging Face |
This project is licensed under the Apache License 2.0.
- Hugging Face for hosting the models
- All the model creators for their incredible work