Welcome to Amphion

Amphion is an advanced open-source toolkit for audio, music, and speech AI research and applications.

Key Features

Text-to-Speech

Convert text to natural-sounding speech with advanced AI models.

Singing Voice Conversion

Transform vocal performances while preserving lyrics and melody.

Text-to-Audio

Generate realistic audio from textual descriptions.

Amphion v2.0 Released

2024-11-15

Major update with improved performance and new features.

New Tutorial: Getting Started with TTS

2024-11-01

Learn how to create your first Text-to-Speech model with Amphion.

Join the Amphion community and start building amazing audio AI applications today!