Simple easy-to-use ChatTTS Dockerfile with GPU Support

This repository provides a Dockerized setup for running ChatTTS, a text-to-speech model designed for dialogue applications, with GPU support. The setup is optimized for ease of use and performance.

Prerequisites

1. Install Docker on Ubuntu 22.04
   Follow the guide to install Docker:
   How to Install and Use Docker on Ubuntu 22.04

2. Perform Post-Installation Steps for Docker
   Ensure you complete the post-installation steps as outlined here:
   Post-Installation Steps for Docker on Linux

3. Install NVIDIA Container Toolkit (For GPU Users)
   If you're using a GPU, install the NVIDIA Container Toolkit by referring to:
   NVIDIA Container Toolkit Installation Guide

4. Install Docker Compose on Ubuntu 22.04
   Set up Docker Compose using the instructions here:
   How to Install and Use Docker Compose on Ubuntu 22.04

Quick Start

Pull the Docker Image

You can pull the pre-built Docker image from Docker Hub:

docker pull naren200/chatts-dockerfile:v1

Run the Docker Container

1. Clone this repository:

   git clone https://github.com/your-username/chatTTS-dockerfile.git
   cd chatTTS-dockerfile

2. Start the Docker container:

   ./start_docker.sh

   This script will:

   • Start the container in attached mode.

3. Once inside the container, you can run the talk.py script to generate speech:

   python3 talk.py

   The output audio files will be saved as basic_output0.wav, basic_output1.wav, etc., in the working directory.

Modify the Text for Speech Synthesis

To change the text that is converted to speech, edit the talk.py file:

1. Open chaTTS-dockerfile/talk.py in your preferred text editor.

2. Modify the inputs_en string with your desired text.

3. Save the file, the saved file will automatically be reflected inside the docker image. Rerun the script:

   python3 talk.py

Stop the Docker Container

To stop and remove the Docker container, run:

./stop_docker.sh

Customizing the Docker Setup

If you want to build the Docker image from scratch or modify the setup:

1. Edit the Dockerfile to add or remove dependencies.

2. Build the Docker image:

   docker build -t naren200/chatts-dockerfile:v1 .

3. Update the docker-compose.yml file if needed (e.g., to change volume mounts or environment variables).

Notes

• The talk.py script is preconfigured to use GPU acceleration. Ensure your system has a compatible NVIDIA GPU.
• The inputs_en string in talk.py supports special tags like [uv_break] for pauses and [laugh] for laughter. Refer to the ChatTTS documentation for more details.
• If you encounter issues with torchaudio, try switching between the two torchaudio.save lines in talk.py.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Star History