Coqui TTS

Coqui TTS is an open-source, deep-learning toolkit providing high-quality, real-time text-to-speech synthesis with advanced voice cloning and multi-language support (17+ languages via the XTTS model).

Coqui TTS delivers a robust, open-source solution for neural Text-to-Speech (TTS), forking from the original Mozilla TTS engine. It leverages state-of-the-art models, notably the XTTS architecture, to transform text into natural-sounding audio instantly (sub-200ms latency). Key features include rapid voice cloning from minimal audio samples (as short as 3 seconds) and extensive multilingual support, currently covering over 17 languages including English, Spanish, and French. This toolkit targets developers and researchers needing customizable, high-volume TTS processing and local deployment capabilities.

https://coqui-tts.readthedocs.io/en/latest/