Multi-Model TTS Service

A unified Text-to-Speech service supporting OpenAI, Gemini, and Local (Coqui TTS/XTTS) models. Built with FastAPI.

Features

• Unified API: Switch between providers (OpenAI, Gemini, Local) with a simple JSON parameter.
• Local Model Support: run privately using Coqui XTTS (supports voice cloning and multi-language).
• FastAPI: reliable, high-performance Python web framework.

Installation

1. Clone the repository

   git clone https://github.com/akashdeepiim/tts.git
   cd tts-service

2. Install Dependencies

   pip install -r requirements.txt

   Note: For local TTS support, ensure you have the necessary system libraries (e.g., espeak).

3. Configuration Create a .env file in the root directory:

   OPENAI_API_KEY=your_openai_key
   GEMINI_API_KEY=your_gemini_key
   USE_GPU=True # Set to False if you don't have a GPU/Metal

Usage

Start the server:

python app/main.py

The API will run at http://localhost:8000.

Generate Audio

Endpoint: POST /v1/tts/generate

Example (OpenAI):

curl -X POST http://localhost:8000/v1/tts/generate \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello world!",
    "provider": "openai",
    "voice_id": "alloy"
  }' \
  --output output.mp3

Example (Local):

curl -X POST http://localhost:8000/v1/tts/generate \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello running locally.",
    "provider": "local",
    "voice_id": "default"
  }' \
  --output local.wav

List Models

Endpoint: GET /v1/tts/providers

curl http://localhost:8000/v1/tts/providers

Documentation

Interactive API documentation is available at http://localhost:8000/docs.