Voice Desk TTS

Voice Desk TTS is a tool that lets you use text to speech voices right in your web browser. It's designed to make it easy to try out different voices without needing complicated setups. You can quickly compare how different voices sound and keep a history of the ones you generate. It also connects with the official list of Piper voices.

Benefits

Voice Desk TTS allows you to run Piper voices directly in your browser. This means you can test and compare voices easily and quickly. It helps you keep track of your generated speech history and syncs with a large catalog of official Piper voices. The tool is lightweight and offers a simple way to manage your text to speech needs.

Use Cases

This tool is useful for anyone who wants to experiment with text to speech voices locally. It's great for comparing different voice options, especially for Vietnamese, English, and Indonesian languages, as these have the strongest browser support. You can also use it to train or fine-tune your own voices using a provided workflow. The generated speech history can be used for quick checks and to see how voices have changed over time.

Vibes

The project is built with modern web technologies like React 19 and Vite, and uses ONNX Runtime Web for efficient voice generation within the browser. It aims to provide a smooth user experience for managing and using TTS voices.

Additional Information

Voice Desk TTS includes a workflow for training new voices, with guides available for English and Vietnamese. After training, the voice files can be easily tested in the browser. The main application code is licensed under Apache-2.0, but users need to be aware of the licensing for the voice assets themselves, including any requirements for attribution or commercial use. The project is not meant to be a complete replacement for all Piper features but focuses on providing a robust browser-based experience.