VoiceLens documentation
Everything you need to grade voice calls with VoiceLens: how the analysis works, how to send calls in, how to read a scorecard, and how to integrate the results into your own systems.
VoiceLens grades every voice call in your call-center, whether an AI agent or a human agent was on the line. Each call gets a health score built from detected faults, per-turn latency, slot capture, question coverage and sentiment, with every finding pinned to the transcript turn it came from.
Start here
Create a workspace, upload a recording or a transcript, and read your first scorecard in a few minutes.
Reading a scorecardWhat the health score, the fault list, the latency chart and the slot table are telling you.
Fault codesThe thirteen faults VoiceLens detects, what triggers each one and how it affects the score.
Rubric templatesDefine the slots to capture, the questions to ask and the phrases to say, per call type.
Integrate
Send recordings straight from your telephony platform with a channel ingest token.
WebhooksReceive completed analyses and threshold breaches with HMAC-signed deliveries.
Embedded dashboardsPut a themed, read-only dashboard in your own portal with one script tag.
Read-only API keysPull conversations and analytics into your own reporting with a workspace API key.
How the documentation is organised
- Getting started walks through the product once, end to end.
- Concepts explains what each signal means and how the health score is computed.
- Guides cover one feature each: channels, templates, calibration, alerts, embeds, sharing, workspaces and billing.
- API reference documents every customer-facing endpoint, header and payload.
- Resources hold the security overview, frequently asked questions and the changelog.