Trust Notice: Voice Sentinel provides probabilistic analysis. Always combine SentinelScore results with external evidence for high-stakes decisions.

Welcome back

Your audio forensics hub. Scan, compare, and explore.

How It Works

1

Record or Upload

Head to the Voice Lab and record live audio from your mic, or upload a WAV or MP3 file from your device.

2

AI Analysis

Our ensemble of deep learning models (CNN, LSTM, TCN, Random Forest, and TSSD) analyse acoustic features to determine authenticity.

3

Get Your Verdict

See a clear Real or Synthetic verdict with a confidence score, model votes, and the full spectral feature breakdown.

About Voice Sentinel

Voice Sentinel is an AI-powered audio forensics platform designed to detect synthetic and deepfake speech. Whether you are a journalist verifying a source, a security professional auditing calls, or simply curious about an audio clip, we give you the tools and transparency to know what is real.

Built with multi-model ensemble analysis. Supports WAV and MP3. Works in your browser.

Voice Lab

Record or upload audio to scan for deepfakes.

Quick flow: Record or upload, review, submit, then confirm whether the verdict looks right.

Ready to Scan

Record or upload an audio file to begin.

AUDIO WAVEFORM VISUALIZATION
Tap Record to start and see the live waveform

Upload Audio

Choose a file from your device and send it straight to review.

Settings

Manage your account, app behavior, and exported data from one place.

Account overview

Guest user

Sign in to sync history and exports.

Basic

Appearance

Dark and light mode

Dark mode
Input device Default
Audio quality High

Data Tools

Exports, service status, and cleanup

Voice Sentinel v1.0

Change user type

Select how you use Voice Sentinel.

Select your user type

Comparative Analysis

Use fresh uploads for new audio, or compare saved results from your history with stored sample IDs.

Rule of thumb: Upload fresh audio with the upload flow. Revisit stored analyses with the saved-samples flow.

Sample 1

Click to upload WAV, MP3, FLAC, or OGG

Sample 2

Click to upload WAV, MP3, FLAC, or OGG

Audio Breakdown

See the verdict and explore the forensic details of your audio.

SentinelScore: Confidence is shown as a percentage estimate, not a certainty.

No analysis selected

Run a scan in Voice Lab, open one from History, or fetch a sample below to start exploring the forensic details.

My Samples

Choose from samples already attached to your account.

Open by Sample ID

Fetch any sample directly when you already know its ID.

History

A compact view of your recent samples and server-side analysis history.

0 History entries

No analysis history yet

Run a scan in Voice Lab and your recent samples and server history will appear here automatically.

Send to Server

Upload a file directly to the backend endpoint.

Any file type is accepted for upload.

Download mobile app

Get Voice Sentinel on your Android device.

Voice Sentinel for Android

Install the app on your phone to record and analyze audio on the go. Same features, full deepfake detection.

Download APK

Requires Android 5.0 or later. You may need to allow installation from unknown sources.