Fish Speech 1.4: open source TTS with voice cloning from 10 seconds and 8 languages
In one sentence Fish Speech 1.4 clones voices from 10s of audio, supports 8 languages, runs real-time on CPU, and offers a serious free alternative to ElevenLabs for developers.
Fish Speech is an open source TTS system that allows cloning any voice from just 10 seconds of sample audio, without expensive GPUs or internet connection. It supports 8 languages including English, Chinese, Japanese, Korean, French, German, Arabic and Spanish, with natural voice quality in all of them. The most interesting thing for application developers is that it runs at real-time speed even on a regular CPU, making it practical for edge devices and offline desktop apps. It is effectively a free alternative to ElevenLabs for those with technical needs who do not want to depend on paid APIs.
Companies
Fish Audio
Tools
Fish Speech
Tags
Sources