Voice-clone defence without magic claims

Audio Forensics & Call Authenticity Lab

Measure audio structure locally, then combine it with identity challenges and scam behavior. The safest question is not “does it sound real?” but “can the caller prove identity without controlling the verification?”

Local execution · source content not transmitted · JavaScript / browser worker—not a WebAssembly claim

Upload call audio or voice note

Stays in your browser

Call authenticity checklist

They created urgency, fear, arrest, accident or emergency pressureThey told me not to contact family, bank staff, employer or policeThey requested money, crypto, gift cards or a transfer to a “safe” accountThey requested screen sharing, AnyDesk, TeamViewer or remote accessThey failed or refused our pre-agreed family/work trust phraseThey repeatedly refused a live video call or independent callbackThe call came unexpectedly or from an unfamiliar numberThey used personal details to sound convincing

Optional exact transcript

Audio cue laboratory

Waveform, clipping, silence, discontinuity and energy cues

Private trust-phrase and callback protocol

Immediate stop/payment protection plan

Audio artifacts are weak testimony. Identity-challenge failure plus a money or secrecy request is much stronger evidence.

Use this check carefully

Inspect a recording you are authorized to analyze and review measurable waveform observations alongside the caller’s request. Use an independently obtained contact number to verify identity.

Interpret the evidence

Clipping, silence and energy changes occur in ordinary recordings too. They are not reliable proof of a synthetic voice. Never approve a payment solely because a familiar voice or a clean waveform seems convincing.