Skip to content

Latest commit

 

History

History
181 lines (137 loc) · 4.21 KB

README.md

File metadata and controls

181 lines (137 loc) · 4.21 KB

MeloTTS API Server

MeloTTS API Server provides an interface to generate high-quality text-to-speech (TTS) audio using the MeloTTS model. The server exposes a RESTful API, allowing users to convert text into speech with support for all base configurations such as speaker voice, speed, and sampling rate.

Features

  • Streamlined REST API for TTS conversion.
  • Support for multiple speakers and languages.
  • Adjustable parameters for speech speed, noise, and sampling rate.
  • Real-time processing for fast and efficient audio generation.
  • Easy integration with client applications.

Installation

Requirements

  • Python 3.9 or higher
  • Required libraries (managed via requirements.txt)
  • Dependencies for audio processing:
    • libsndfile1 (Linux/Unix)
    • FFmpeg (if using pydub for audio manipulation)
  • Docker (optional, for containerized deployment)

Steps

  1. Clone the repository:

    git clone https://github.com/nyedr/MeloTTS_Server_Api.git
cd melotts-api-server

  2. Install required dependencies:

    pip install -r requirements.txt

  3. Download necessary linguistic resources and pre-trained models:

    python -m unidic download
python melo/init_downloads.py

  4. Run the server:

    python main.py

  5. The server will start on http://127.0.0.1:8000 by default.

Usage

Endpoints

1. /tts/generate (POST)

Generates a speech audio file from the given text.

  • URL: http://127.0.0.1:8000/tts/generate
  • Method: POST
  • Headers: Content-Type: application/json
  • Request Body: 
    {
  "text": "Your text to convert to speech.",
  "voice_id": "EN-US",  
  "sr": 22050,          
  "speed": 1.0          
}
  • Response: Streams the audio file in WAV format.

2. /speakers (GET)

Fetches the list of available speaker IDs.

  • URL: http://127.0.0.1:8000/speakers
  • Method: GET
  • Response: 
    {
  "available_speakers": ["EN-US", "EN-BR", "EN-AU", "EN-INDIA", "EN-Default"]
}

Examples

Generating Speech

Using curl

curl -X POST "http://127.0.0.1:8000/tts/generate" \
    -H "Content-Type: application/json" \
    -d '{
          "text": "Hello, this is a test of the MeloTTS API.",
          "voice_id": "EN-US",
          "sr": 22050,
          "speed": 1.0
        }' --output output.wav

Using Python

import requests

url = "http://127.0.0.1:8000/tts/generate"
data = {
    "text": "Hello, this is a test of the MeloTTS API.",
    "voice_id": "EN-US",
    "sr": 22050,
    "speed": 1.0
}
response = requests.post(url, json=data, stream=True)

with open("output.wav", "wb") as f:
    for chunk in response.iter_content(chunk_size=1024):
        if chunk:
            f.write(chunk)

Docker Deployment

Building the Image

docker build -t melotts-api-server .

Running the Container

docker run -p 8000:8000 melotts-api-server

Configuration

Environment Variables

You can configure the server behavior using the following environment variables:

Variable Default Value Description
HOST 0.0.0.0 Server host address.
PORT 8000 Server port.
DEFAULT_SPEED 1.0 Default speech speed.
DEFAULT_LANGUAGE EN Default language for the TTS model.
DEFAULT_SPEAKER_ID EN-US Default speaker voice ID.

Contributing

Contributions are welcome! Please fork the repository and submit a pull request with your changes.

  1. Clone the repository:

    git clone https://github.com/nyedr/MeloTTS_Server_Api.git

  2. Create a feature branch:

    git checkout -b feature-name

  3. Commit your changes and push:

    git add .
git commit -m "Add new feature"
git push origin feature-name

  4. Open a pull request on GitHub.

Fork of MeloTTS