Question 1

How accurate is the audio deepfake detector?

Accepted Answer

aigeneratedit's audio detector achieves 96% accuracy on the ASVspoof 2021 benchmark using an ensemble of Wav2Vec2, CNN-BiLSTM, and MFCC-CNN models.

Question 2

What audio formats are supported?

Accepted Answer

We support MP3, WAV, M4A, OGG, FLAC, and AAC. Maximum file size is 50MB on the free plan.

Question 3

Can it detect voice cloning from tools like ElevenLabs?

Accepted Answer

Yes. Our ensemble is trained on synthetic audio from ElevenLabs, Resemble.ai, VALL-E, and other state-of-the-art TTS systems.

Question 4

How long does an audio scan take?

Accepted Answer

Most audio files under 10 minutes return results in under 8 seconds. Pro users get priority queue access with average latency under 3 seconds.

Question 5

Is my audio stored or used to train AI models?

Accepted Answer

No. Your files are stored temporarily on encrypted Cloudflare R2 storage and automatically deleted within 24 hours. We never use uploaded content to train our models.

Audio Deepfake Detector

Frequently asked questions

Integrating into a workflow?