Audio_Analysis-API

This project is a FastAPI-based backend that analyzes 5–10 audio files to extract speech features and calculate a cognitive risk score. It supports file uploads, processes audio using custom utilities, and returns results via a REST API.

---

📜 Project Overview

This API processes uploaded audio files, extracts key speech features (MFCC, ZCR, RMS), and calculates a cognitive risk score. The application is built using FastAPI and deployed on Render.

---

🛠 Installation

Requirements:

• Python 3.7 or higher
• pip (Python package installer)
• system dependencies like ffmpeg for audio processing

Steps:

1. Clone the repository:

   git clone https://github.com/your-username/Audio_Analysis-API.git
   cd Audio_Analysis-API

2. Create a virtual environment:

   python -m venv venv
   source venv/bin/activate  # For Linux/MacOS
   venv\Scripts\activate     # For Windows

3. Install dependencies:

   pip install -r requirements.txt

4. Install system dependencies (for audio processing):

   • If you're using librosa for audio processing, you may need ffmpeg or libsndfile. You can install ffmpeg with:

     sudo apt-get install ffmpeg  # For Ubuntu
     brew install ffmpeg          # For MacOS

---

⚙️ Usage

Running Locally:

To run the FastAPI server locally, use the following command:

uvicorn app.main:app --reload