Text-to-Speech System
39.1k 2026-04-18
2noise/ChatTTS
A generative speech model optimized for natural and expressive daily dialogue, especially for LLM assistants.
Core Features
Conversational TTS optimized for dialogue scenarios and multiple speakers
Fine-grained control over prosodic features like laughter, pauses, and interjections
Superior prosody compared to most open-source TTS models
Supports English and Chinese languages
Provides pretrained models for academic research and development
Quick Start
git clone https://github.com/2noise/ChatTTS && cd ChatTTS && pip install --upgrade -r requirements.txtDetailed Introduction
ChatTTS is an advanced text-to-speech (TTS) model specifically engineered for dialogue scenarios, such as those involving LLM assistants. It excels in generating natural and expressive speech, supporting multiple speakers for interactive conversations. A key highlight is its ability to predict and control fine-grained prosodic features like laughter, pauses, and interjections, leading to superior prosody compared to many open-source alternatives. The project provides pretrained models, trained on extensive Chinese and English audio data, to foster further research and development in conversational AI.