Subscribe to get weekly email with the most promising tools 🚀

Optimized for natural, conversational text-to-speech

Listed in categories:



ChatTTS is a text-to-speech tool optimized for natural and conversational scenarios. It is designed for dialogue tasks of large language model assistants and applications requiring conversational audio and video introductions. ChatTTS supports both Chinese and English languages and demonstrates high quality and naturalness in speech synthesis.

How to use ChatTTS?

To use ChatTTS, follow these steps: 1. Download the code from GitHub. 2. Install necessary dependencies like torch and ChatTTS. 3. Import required libraries in your script. 4. Initialize ChatTTS and load the pre-trained models. 5. Prepare the text you want to convert to speech. 6. Generate speech using the infer method. 7. Play the generated audio using the Audio class from IPython.display.

Core features of ChatTTS:


Multi-language Support


Large Data Training


Dialog Task Compatibility


Open Source Plans


Control and Security


Ease of Use

Why could be used ChatTTS?

#Use caseStatus
# 1Conversational tasks for large language model assistants
# 2Generating dialogue speech
# 3Video introductions

Who developed ChatTTS?

ChatTTS is developed by a dedicated team committed to improving the controllability, security, and naturalness of the model. The team aims to open-source a trained base model to facilitate further research and development within the community.

FAQ of ChatTTS