Home · Detect · Descript

Detect Descript AI voices.

Updated July 2026

Suspect a clip was made with Descript? Drop it in and get a citable verdict with the model named, in under half a second. Editor-native voice cloning (Overdub) plus stock AI voices, tuned to blend into a real recorded track.

Synthetic detectedDescript
Confidence
high
Model
Descript
Yes, you can check. Drop the clip into the detector and it returns the probability that the speech was generated by Descript, a confidence level, and the named model, in under half a second. It reads the audio's synthesis signature, so it works on cloned and stock Descript voices alike, and gives you a citation URL to quote.
What it is

Descript, and why it shows up in suspicious audio

Descript is a widely used audio and video editor whose Overdub feature clones a voice from a training sample, letting an editor fix or add speech by typing rather than re-recording. It is a staple of podcast and video production.

Because Overdub is built to blend seamlessly into a genuine recording, an unlabeled synthesized insert can sit inside otherwise-real audio, which is exactly the case worth checking.

Where you tend to see it: Podcast and video edits, corrections spliced into real recordings, and cloned-voice inserts.

Descript is a legitimate product; misuse is the problem, not the tool. You can read about it on the official Descript site.

The tells

How to tell a Descript voice

The human ear is unreliable on current Descript audio. These are the signals a detector weighs. We report which ones drove the verdict rather than handing you a bare number.

  • 01Seams where a cloned insert meets real recorded audio, heard as small shifts in room tone or level.
  • 02Prosodic evenness in the synthesized segments that the surrounding human speech does not share.
  • 03Timbre that stays fixed across an insert where a re-recorded take would drift.
  • 04Vocoder artifacts in fricatives that survive the editor's export and re-encoding.
Spectral view · artifacts concentrate where synthesis smooths what a human voice would not
How the detector identifies Descript

A verdict you can cite, not a vibe

The detector reads the audio, not the speaker. Cloning or stock voice, clean or compressed, it looks for the synthesis signature and attributes the source.

Step 1

Drop the clip

Upload a file or paste a URL. MP3, WAV, M4A, WebM, or the audio track of a video. About half a second of clear speech is enough.

Step 2

The model scores it

The same model behind the public detector reads the acoustic signature and weighs the artifacts, then attributes the source, e.g. Descript, when it recognizes it.

Step 3

Get a citable verdict

You get a probability, a confidence level, the named model, and a permanent citation URL you can quote, file, or subpoena.

0.48s
Median verdict
99%
Accuracy on clean audio
24+
Generators covered
24h
Audio deleted after
If a clip turns out synthetic

What to do next

A verdict is evidence, not a verdict of intent. Save the result to get a permanent citation URL and a one-way audio fingerprint you can reference later without storing the file. If you are a journalist or investigator, cite the verdict alongside your own reporting; if this is a suspected scam, treat the contact as unverified and confirm through a channel you already trust. The FTC's advice on suspected scam calls is a sensible baseline.

Building this into a workflow? The API returns the same verdict as JSON with webhooks for bulk jobs, and the browser extension checks audio in place on WhatsApp Web, YouTube, and podcasts.

FAQ · Descript

Common questions

Can you detect a Descript Overdub insert inside a real recording?
Often, yes. The detector reads the synthesis signature, so a cloned insert can stand out even when it is spliced into genuine audio, though short inserts lower confidence.
Does it work on a whole edited episode?
Yes, though a long file with only a brief synthetic insert is harder. Isolating the suspect segment gives a clearer verdict.
Is a single check free?
Yes, a single verdict is free with no card, showing the probability, confidence, and recognized model.
Will it name Descript specifically?
When the signature is recognizable we name it; otherwise we return 'unknown synthesis' rather than guess a brand.
What happens to my audio?
It is deleted within 24 hours. Saved verdicts keep only a one-way fingerprint, never the file.

Is this a Descript voice? Find out.

Free verdict, model named, in under a second. No card to start.

Open detectorUse the API