ChatTTS is a voice generation model for conversational scenarios in Chinese and English.
What is ChatTTS?
ChatTTS is a voice generation model specifically designed for dialogue scenarios. It is highly suitable for use in dialogue tasks associated with large language model assistants, as well as in the creation of dialogue-based audio and video introductions. This model supports both Chinese and English languages and delivers high-quality, natural-sounding output during speech synthesis. Such excellent performance is achieved through training using approximately 100,000 hours of data in both Chinese and English. The project team plans to make the base model, which was trained using 40,000 hours of data, available open-source, in order to assist academic and development communities in conducting further research and innovation.
How to use ChatTTS?
To use ChatTTS, please download the code from GitHub, install the necessary dependencies (torch and ChatTTS), import the required libraries, initialize ChatTTS, prepare the text, generate the audio using the `infer` method, and play the generated audio using the `Audio` class in `IPython.display`.
The core functions of ChatTTS
Multi-language support (English and Chinese)
High-quality voice synthesis with a natural listening experience
Compatible with dialogue tasks using large language model assistants
The open-source initiative offers pre-trained basic models.