Detect Coqui AI voices.
Suspect a clip was made with Coqui? Run it through the free detector to check whether the voice is AI-generated. Coqui in short: open-source text-to-speech with multilingual voice cloning (XTTS) from a short sample.
Coqui, and why it shows up in suspicious audio
Coqui is a widely used open-source text-to-speech project whose XTTS models can clone a voice in many languages from a few seconds of audio. Because it is free and self-hostable, it turns up in hobbyist projects, indie tools, and occasionally impersonation, with no platform gatekeeping.
Open-source also means there is no account trail to follow: anyone can run it locally, so the audio itself is the only evidence, which is exactly where a detector earns its place.
Where you tend to see it: Self-hosted projects, indie apps, and cloned-voice impersonation.
Coqui is a legitimate product; misuse is the problem, not the tool. You can read about it on the official Coqui site.
How to tell a Coqui voice
The human ear is unreliable on current Coqui audio. These are the kinds of artifacts a detector is trained to pick up.
- 01Multilingual cloning artifacts that a model reads even when the language sounds native.
- 02Prosodic evenness across sentences a human speaker would vary.
- 03Timbre that stays fixed where a real voice would drift.
- 04Vocoder artifacts in fricatives that can survive re-encoding.
What the detector tells you
The detector reads the audio, not the speaker. Cloned or stock voice, clean or compressed, it looks for the signs of synthesis.
Drop the clip
Upload an MP3, WAV, M4A, OGG, FLAC or AAC file. Three seconds or more of clear speech gives the most reliable result.
The model scores it
The model scores the clip. On longer recordings it scores each segment separately, so you can see which part sounds synthetic.
Read the verdict
You get a verdict on a four-step scale, from Human to AI Generated, plus an AI probability score. It tells you whether the voice is AI-generated, not which tool made it.
What to do next
A verdict is evidence, not proof. Sign in to keep the result in your detection history and share it with a link. If you are a journalist or investigator, use it 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, and a Chrome extension for checking audio in any tab is in development.