ChatTTS
Conversational text‑to‑speech model for natural Chinese and English voice output
ChatTTS is an open‑source voice generation model optimized for dialogue scenarios. Trained on ~100,000 hours of Chinese and English speech, it delivers high‑quality, natural‑sounding audio for LLM assistants, video introductions, educational content, and any app needing text‑to‑speech. Developers can integrate via API/SDK, fine‑tune for custom voices, and access a planned base model trained on 40,000 hours for research and further development.