Description
You want voice cloning you host yourself, without being tied to a cloud API's quota and pricing. Fish Speech is one of the most popular open-source TTS and voice cloning projects, with good multilingual output, emotional expression and streaming generation, over thirty thousand stars, and you run it on your own machine or server.
It ships a command line, an API server and a WebUI, with Docker deployment and clear official docs for installation, weights and inference. Mind the license: the code and associated weights fall under the Fish Audio Community License and the S2-Pro weights under the Fish Audio Research License, so read the LICENSE before commercial use.
Voice cloning: Clone a voice from reference audio with natural emotion
Multilingual: Speech synthesis across many languages
Streaming generation: Audio plays as it generates, suited to real-time dialogue
Several ways to run: Command line, API server and WebUI
Docker deployment: Ready images and official docs for installation, weights and inference
SGLang-Omni for efficiency: Recommended by the authors for users who care about speed
It ships a command line, an API server and a WebUI, with Docker deployment and clear official docs for installation, weights and inference. Mind the license: the code and associated weights fall under the Fish Audio Community License and the S2-Pro weights under the Fish Audio Research License, so read the LICENSE before commercial use.
Features
Voice cloning: Clone a voice from reference audio with natural emotion
Multilingual: Speech synthesis across many languages
Streaming generation: Audio plays as it generates, suited to real-time dialogue
Several ways to run: Command line, API server and WebUI
Docker deployment: Ready images and official docs for installation, weights and inference
SGLang-Omni for efficiency: Recommended by the authors for users who care about speed
