Audio Assistant

Objective

This project is a real-time audio transcription application. It captures audio from the user's microphone via a React frontend, streams it to a Python backend, and uses the faster-whisper library (specifically the large-v3-turbo model) to transcribe the speech into text in real-time.

Project Structure

• frontend/: A React application built with Vite and TypeScript. It handles audio recording, visualization, and displaying the transcription.
• backend/: A Python FastAPI server. It receives audio streams via WebSockets, processes them using ffmpeg and numpy, and performs transcription using faster-whisper.

Prerequisites

• Node.js (v18 or higher)
• Python (v3.8 or higher)
• FFmpeg: Required for audio decoding on the backend.

  sudo apt-get install ffmpeg

Quick Start

1. Start the Backend

cd backend
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn server:app --reload --host 0.0.0.0 --port 8000

2. Start the Frontend

Open a new terminal:

cd frontend
npm install
npm run dev

3. Usage

• Open your browser to http://localhost:5173.
• Click "Mic OFF" to enable the microphone.
• Click "Record" to start streaming audio to the backend.
• Speak clearly. The transcription will appear in the text box below the visualizer.
• Click "Stop Recording" to end the session.
• Click "Play Demo" to replay your recorded audio.

Running Independently

The frontend and backend are decoupled and run as separate processes.

• Backend: Runs on port 8000. Handles the WebSocket connection and transcription logic.
• Frontend: Runs on port 5173. Serves the UI and connects to the backend.

You can start, stop, or restart either service without affecting the other (though functionality will be limited if one is down).