Skip to content

DOCKER.md

Latest commit

 

History

History
114 lines (76 loc) · 2.82 KB

DOCKER.md

File metadata and controls

114 lines (76 loc) · 2.82 KB

Docker Usage Guide for PyThaiTTS

This guide explains how to build and run PyThaiTTS using Docker.

Building the Docker Image

To build the Docker image, run the following command from the root directory of the repository:

docker build -t pythaitts:latest .

This will create a Docker image named pythaitts:latest with all dependencies installed.

Running the Demo

To run the demo script that demonstrates Thai text-to-speech synthesis:

docker run --rm pythaitts:latest

The demo will:

  1. Initialize the PyThaiTTS model (default: lunarlist_onnx)
  2. Generate speech from Thai text
  3. Save the output to a WAV file
  4. Display the waveform information

Custom Usage

Interactive Shell

To start an interactive shell inside the container:

docker run --rm -it pythaitts:latest /bin/bash

Run Custom Python Script

To run your own Python script:

docker run --rm -v $(pwd)/your_script.py:/app/custom.py pythaitts:latest python custom.py

Save Output Files

To save generated audio files to your host machine:

docker run --rm -v $(pwd)/output:/app/output pythaitts:latest python -c "
from pythaitts import TTS
tts = TTS()
tts.tts('สวัสดีครับ', filename='output/hello.wav')
"

This will save the generated hello.wav file to the output directory on your host machine.

Example Usage in Python

Inside the container, you can use PyThaiTTS as follows:

from pythaitts import TTS

# Initialize TTS with default model
tts = TTS()

# Generate speech and save to file
file_path = tts.tts("ภาษาไทย ง่าย มาก มาก", filename="output.wav")
print(f"Audio saved to: {file_path}")

# Generate speech and get waveform
waveform = tts.tts("ภาษาไทย ง่าย มาก มาก", return_type="waveform")
print(f"Waveform shape: {waveform.shape}")

Available TTS Models

PyThaiTTS supports multiple models:

  • lunarlist_onnx (default): ONNX-optimized model, CPU-only
  • khanomtan: KhanomTan TTS model
  • lunarlist: Original Lunarlist model

To use a different model:

from pythaitts import TTS

# Using KhanomTan model
tts = TTS(pretrained="khanomtan", version="1.0")

Requirements

  • Docker installed on your system
  • At least 2GB of available disk space
  • Internet connection for downloading models on first run

Troubleshooting

If you encounter issues with model downloads, ensure:

  1. You have a stable internet connection
  2. The Hugging Face Hub is accessible from your network
  3. You have sufficient disk space for model files

Notes

  • The first run will download model files from Hugging Face Hub, which may take some time depending on your internet connection
  • Generated audio files are in WAV format
  • The default model (lunarlist_onnx) runs on CPU and doesn't require GPU support