AI Voice Detector

What it does

AI Voice Detector analyzes audio and video clips to estimate whether speech is human or synthetic. It returns a probability with confidence scores, names the generator when it recognizes one, and gives each verdict a permanent URL and APA-style citation. The same model runs on every plan, from Free to Enterprise.

AI Voice Detector is a web tool for checking whether a recording sounds human or synthetic. You drop in an audio file, paste a URL, or record from the browser, and it returns a verdict in about half a second. The result includes a probability, a confidence score, and, when the generator is recognized, a model name such as ElevenLabs or OpenAI. Every verdict gets a permanent URL and an APA-style citation, which helps when you need to reference a result in editorial or legal work. The site says the model is retrained monthly and that accuracy is 99% on clean audio, with lower confidence expected on compressed audio like phone calls. The Free plan allows 50 detections per month with clips up to 2 minutes, and no card is required. Paid plans add more detections, longer clips, PDF and CSV export, and API access. The site does not state that the code is open source, so self-hosting is not confirmed.