
Text-to-speech for conversational scenarios — a voice generation model optimized for natural, conversational TTS.
ChatTTS is a voice generation model designed specifically for conversational scenarios — the dialogue tasks of large language model (LLM) assistants, plus conversational audio and video introductions. Trained on approximately 10 million hours of Chinese and English data, it delivers high quality and naturalness in speech synthesis.
It supports both Chinese and English, is well-suited for dialog tasks, and the project team plans to open source a trained base model. The team is committed to improving controllability, adding watermarks and integrating with LLMs for safety and reliability.
With 20K+ GitHub stars, ChatTTS is easy to use — just input text and generate corresponding voice files. A free online demo and free AI voice cloning (via Voicv) are available.
Developers, AI researchers, Content creators, App builders, Voice AI enthusiasts
AI-powered voice cloning platform that replicates, synthesizes, and transcribes speech with zero-shot learning and multilingual support.
WhisperUIPaidSpeech to text with OpenAI Whisper — turn audio, meetings, and recordings into usable text.
YescribeFreeFast, accurate, and secure AI-powered transcription service that converts audio and video to text in 98 languages, with 99.9% precision and up to 5-hour upload support.
TextPixieFreeTextPixie is a free online AI translator that instantly converts text, images, audio, documents, and web articles into over 100 languages with fast, accurate, context-aware results for personal and professional communication.