NeuTTS-FastAPI

A FastAPI wrapper for NeuTTS-Air, providing an OpenAI-compatible text-to-speech API.

Features

• OpenAI-Compatible API: Drop-in replacement for OpenAI's TTS API
• CPU-Only Inference: Optimized for CPU deployment
• FastAPI Backend: High-performance async API server
• Docker Support: Easy containerized deployment
• Voice Cloning: Support for custom voice models

Quick Start

Using Docker (Recommended)

1. Clone the repository:

   git clone https://github.com/hasithdd/NeuTTS-FastAPI.git
   cd NeuTTS-FastAPI

2. Build and run with Docker Compose:

   cd docker/cpu
   docker-compose up --build

3. Test the API:

   curl -X POST "http://localhost:8000/v1/audio/speech" \
     -H "Content-Type: application/json" \
     -d '{
       "input": "Hello, world!",
       "voice": "dave"
     }' \
     --output speech.wav

Local Development

1. Install dependencies:

   pip install uv
   uv sync

2. Run the server:

   uv run uvicorn api.src.main:app --host 0.0.0.0 --port 8000

API Usage

POST /v1/audio/speech

Generate speech from text using the specified voice.

Request Body:

{
  "input": "Text to convert to speech",
  "voice": "dave"
}

Supported Voices:

• dave: Male voice
• jo: Female voice

Response:

• Content-Type: audio/wav
• Body: WAV audio data

Example:

curl -X POST "http://localhost:8000/v1/audio/speech" \
  -H "Content-Type: application/json" \
  -d '{"input": "Hello, world!", "voice": "dave"}' \
  --output hello.wav

Project Structure

NeuTTS-FastAPI/
├── api/
│   └── src/
│       ├── main.py              # FastAPI application
│       ├── routers/
│       │   └── openai_compatible.py  # TTS endpoint
│       └── voices/              # Voice model files
│           ├── dave.pt
│           └── jo.pt
├── neutts-air/                  # NeuTTS-Air library
├── docker/
│   └── cpu/                     # Docker configuration
├── pyproject.toml               # Project dependencies
└── README.md

Dependencies

• Python 3.10+
• NeuTTS-Air
• FastAPI
• PyTorch (CPU)
• SoundFile
• And other dependencies listed in pyproject.toml

Development

Running Tests

uv run pytest

Code Formatting

uv run black .
uv run isort .

Roadmap

Core Features

• GPU Support: Add CUDA acceleration for faster inference on GPU hardware
• Voice Cloning: Enable users to create custom voices from audio samples
• Streaming Audio: Implement real-time audio streaming for low-latency TTS
• Multiple Audio Formats: Support MP3, FLAC, and other popular audio formats
• Batch Processing: Allow processing multiple text inputs in a single request

Enhanced Functionality

• Web Interface: Create a user-friendly web UI for testing and demonstration
• Voice Mixing: Combine multiple voices or adjust voice characteristics
• Emotion Control: Add parameters for controlling speech emotion and style
• Multilingual Support: Extend beyond English to support other languages
• API Rate Limiting: Implement request throttling and usage controls

Technical Improvements

• Performance Optimization: Benchmark and optimize inference speed
• Comprehensive Testing: Add unit tests, integration tests, and CI/CD pipeline
• API Documentation: Generate interactive OpenAPI/Swagger documentation
• Container Optimization: Reduce Docker image size and improve startup time
• Monitoring & Metrics: Add logging, metrics, and health monitoring

Contributing

See Contributing.md for guidelines.

License

MIT License - see LICENSE file for details.

Acknowledgments

• Based on NeuTTS-Air
• Inspired by Kokoro-FastAPI