Fish Speech: Open-source Multilingual TTS with Inline Control and Production-ready Streaming
Summary
Fish Speech S2 is a state-of-the-art open-source text-to-speech system from Fish Audio, offering multilingual support, fine-grained inline control, and production-ready streaming. The repository details architecture, licensing, benchmark results, and how to run or deploy via SGLang, with a focus on high fidelity, multi-speaker generation, and rapid voice cloning.