Real Time Detection

Overview

VoxShield uses browser microphone capture to simulate real-time audio analysis.

The system does not intercept live phone calls or communication streams.

Processing Flow

Browser Microphone
       |
       v
Audio Recording
       |
       v
Client-Side WAV Conversion
       |
       v
Flask API
       |
       v
Feature Extraction
       |
       v
ML Model
       |
       v
Prediction

Why WAV Conversion?

The frontend converts recorded audio into WAV before sending it to the backend.

This avoids requiring the backend to decode browser-generated WebM/Opus audio.

Scope Limitation

The current system is a demonstration of audio classification using microphone input.

It does not establish continuous monitoring or real-time interception of external calls.