TTS-API

Overview

This project provides a FastAPI-based backend for performing the following tasks:

1. OCR (Optical Character Recognition): Extract Arabic text from images.
2. Text-to-Speech (TTS): Convert text to speech and return an audio file.

Endpoints

1. /ocr - Extract Arabic Text from Images

• Method: POST
• Description: Upload an image to extract Arabic text using OCR.
• Request:
  • file (form-data): The image file.
  • lang (form-data, optional): Language code (default: ara).
• Response:
  • Extracted text in JSON format.

3. /tts - Convert Text to Speech

• Method: POST
• Description: Convert text to speech and return an audio file.
• Request:
  • text (form-data): Text to convert to speech.
  • voice (form-data, optional): Voice name (default: Aisha).
• Response:
  • Audio file in MP3 format.

How to Use with Frontend

1. OCR:

   • Use a file input to upload an image.
   • Send a POST request to /ocr with the image file and optional language code.
   • Display the extracted text from the response.

2. Text-to-Speech:

   • Send a POST request to /tts with the text and optional language code.
   • Play or download the returned MP3 file.

Setup

1. Install dependencies:

   pip install -r requirements.txt

2. Run the FastAPI server:

   uvicorn main:app --reload

3. Access the API documentation at http://localhost:8000/docs to test the endpoints interactively.

Prerequisites

Install CUDA

Ensure that CUDA is installed on your computer to enable GPU acceleration for supported libraries.

• Download CUDA from the NVIDIA CUDA Toolkit website.
• Follow the installation instructions for your operating system.