Drop in an audio file or call recording. Detect synthetic speech from ElevenLabs, Resemble, PlayHT, OpenAI, and 50+ other voice-cloning engines.
Synthetic speech is statistically different from human speech. Our model triangulates across six independent signals — the file has to fail multiple before we flag it.
Real speakers vary pitch involuntarily — micro-fluctuations every 80–150ms. TTS systems smooth them out. We measure the variance.
Humans breathe. Cloned voices often skip the inter-clause inhale or insert a synthetic one of suspiciously consistent duration.
Vocoder upsampling leaves a high-frequency quantization signature most TTS engines can't fully suppress — visible at 8kHz+.
Diffusion-based voice clones produce unnaturally smooth phoneme transitions. We compare against a corpus of 200k human speech samples.
Phone calls have background noise. A "phone call" with a studio-clean noise floor is one of the strongest deepfake tells.
Most TTS systems have a unique frequency-response signature. We match against signatures from the 50+ engines we track.
If a TTS or voice-clone product has been used in a documented attack, our detector knows its fingerprint. Models retrained monthly as new engines ship.
Documented voice-clone fraud in the last 24 months. Every one had a clear acoustic signal we'd have flagged inside a second.
A woman in Hyderabad lost Rs 1.4 lakh to a scammer using AI to mimic her nephew's voice. The clone had a flat pitch contour and missing breath gaps — both Signal 01 and 02 flags.
A finance worker was tricked into sending $25M after a deepfake video call with cloned voices of the CFO and colleagues. Studio-clean noise floor on a "Zoom call" — Signal 05 alone would have triggered.
A mother received a fake-kidnapping ransom call using a cloned daughter's voice. The cry-and-plead pattern was tonally consistent across 90 seconds — humans aren't.
Free plan ships with 50 detections/month. Starter at $49/mo handles 1,000. Enterprise scales to 20,000.