Voice Cloning API

This project provides an API for voice cloning using text and voice input files. It leverages a deep learning model for text-to-speech (TTS) synthesis to generate cloned voice from the provided text and voice samples.

Prerequisites

Before getting started, make sure you have the following installed:

• Python >= 3.9, < 3.12..
• A GPU with CUDA (version > 2) support for optimal performance.
• CUDA-enabled PyTorch installation.
• The required Python packages (listed in requirements.txt).

Installation

Follow these steps to set up the project on your local machine:

1. Clone the repository or download the project files to your local machine.

2. Navigate to the project directory where the requirements.txt is located.

3. Install the required dependencies by running:

   pip install -r requirements.txt

4. Start the Flask app on port 8888 by running:

   flask --app main.py run --port 8888

   The application will start running at http://127.0.0.1:8888.

API Endpoints

1. Health Check Endpoint

curl --location 'http://127.0.0.1:8888'

2. Voice Cloning

curl --location 'http://127.0.0.1:8888/tts/voice-cloning' \
  --form 'text=@"/home/arjun/Desktop/tts-api/text/test.txt"' \
  --form 'voice=@"/home/arjun/Desktop/tts-api/audio/test.wav"'