Consent
Every task starts with contact provenance, clear permission, and a stop/delete path before any voice data is requested.
Fixture-backed evidence path
This page shows the current Voice Bench SA pipeline using safe fixture data only. It is a working shape for consent, voice-note intake, transcript correction, independent review, and export evidence.
Every task starts with contact provenance, clear permission, and a stop/delete path before any voice data is requested.
The fake fixture accepts five WhatsApp-style audio submissions across isiXhosa, Afrikaans, South African English, and mixed phrase tasks, while rejecting unsafe or incomplete media before export.
Audio is paired with transcript candidates and participant corrections so local names, speech, and phrasing can be checked.
A separate reviewer validates the correction path before any row can enter the export-ready batch.
The generated manifest has five rows and points at the planned Google Cloud Storage object, with no live cloud write yet.
Export manifest
e03756f6ce3fdfd4327adb95ffdd3b3d523d0835002e85b98fcdeacc0c7732c6
Export gates
npm run xprize:local-proof:check.npm run xprize:benchmark-snapshot:report.Gemini QA contract
Current status