Skip to main content

Solution

Recorded-request review for banking and wire workflows

Sonotheia measures voice and channel properties in recorded requests for banking review, and gives the reviewer a record of what it measured and how valid the audio was.

The intended use is review after the call and before the transfer. An instrument reading adds evidence to existing procedures and does not approve or block a transfer.

Where can a review fit?

Scope the actual pause between a recorded request and a consequential action, such as a wire transfer, credential reset or exception approval. Confirm that the recording and a reviewer are available in that window.

What does the instrument observe?

Sonotheia measures physical properties of the voice and the channel in call audio and reports each one as nominal, deviated or unavailable, with the validity of the recording. A reviewer decides what to do with the result. Sonotheia does not decide whether a voice is synthetic.

What does integration expose?

The analysis endpoint returns a measurement frame: the validity of the audio, voice and channel readings, and a recommended human action. It does not return an identity or fraud determination. Evidence exports are separate from this response.

Frequently asked questions

Does Sonotheia establish regulatory compliance?
A structured measurement record with the instrument version can support a compliance team's own review and SAR narrative preparation. It is not a qualified SAR filing or a determination that a reporting requirement has been met. The institution and its legal reviewers decide what to file.
Can Sonotheia detect a named commercial voice clone?
No generator-specific detection claim is made. Named measurements do not establish whether a voice is synthetic.
Does this work on standard telephony networks?
Sonotheia's measurements are benchmarked on simulated wideband, G.711 and AMR-NB chains built from public spoofing corpora. They have not been validated on live telephone traffic. A real trunk adds handset, network and transcoding effects the simulation did not include, so it stays pending until measured in place. Zoom, Teams and WebRTC meeting audio are out of scope.
What happens when there is too little speech?
Where a recording holds too little speech to read, Sonotheia reports not enough to act. This does not promise a served out-of-distribution gate for every speech-bearing recording.
Is customer audio sent to Sonotheia cloud for training?
Confirm data handling separately for your deployment: audio storage, result retention, access controls, training use and subprocessors before use.
Does this verify liveness?
No. Human origin, identity and authorization remain separate questions.
What is the first step for a bank evaluation?
Discuss the recorded-request workflow, channel mix and acceptance criteria before any production commitment.

Ready to evaluate Sonotheia for your voice channel?

Request a demo