Workflow focus
Sonolock is designed to make voice-agent workflow context reviewable, not to identify who a person is.
SONOLOCK
Sonolock is the Voice Agent Provenance / Disclosure & Accountability View.
Target model: Voice Agent → Session → Authorization Evidence → Action Evidence → Disclosure → Human Escalation → RecordPack. The intended RecordPack organizes voice-agent provenance, disclosure, and human-review context. The public preview is metadata-only; live generation and delivery are not active.
SonolockControlled preview. Metadata-only. Sonolock does not identify speakers, perform voiceprint recognition, process biometric identifiers, authenticate credentials, adjudicate truth or fraud, or make automated decisions. By default, it does not retain raw audio.
Intended review inputs
- Voice-agent sessions
- Authorization evidence
- Action evidence
- Disclosure records
- Human escalation
- Incident records
- Remediation notes
- Review context
Intended record artifacts
- Session provenance summary
- Authorization and action evidence map
- Disclosure record
- Human escalation record
- Incident and remediation record
- Intended reviewable RecordPack representation
- Human review checklist
product / brand surface
Sonolock brand overview
Intended outputs
Intended voice-agent provenance and disclosure context, timeline summary, operating node list, authorization and action evidence, human escalation record, and snapshot-bound RecordPack representation.
Boundary
Sonolock does not identify speaker identity, perform voiceprint recognition, process biometric identifiers, authenticate credentials, or make automated verification or adjudication decisions. By default, it does not retain raw audio.