Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.superbryn.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

​
What is production monitoring?

SuperBryn ingests your agent’s real production calls, analyzes every conversation, and surfaces quality issues automatically. You connect a data source, and SuperBryn handles transcription analysis, latency measurement, sentiment detection, and LLM-powered auditing.

​
Workflow

1

Connect a data source

Link your voice AI provider via webhook, API key, SDK, or upload recordings directly.
2

Calls are ingested

Call data flows into SuperBryn - transcripts, recordings, metadata, and cost information.
3

Automated analysis

Each call is processed for quality metrics, sentiment, latency breakdowns, and interruption patterns.
4

LLM-powered audit

An AI auditor evaluates every call across four dimensions: call path adherence, policy compliance, security, and tool call handling.

​
Supported providers

ProviderIntegration methods
VapiWebhook, API key sync
RetellWebhook, API key sync
ElevenLabsWebhook, API key sync
BlandWebhook, API key sync
LiveKitPython SDK (livekit-evals)

​
Other data sources

SourceDescription
Audio uploadDrag-and-drop audio files (MP3, WAV, M4A, OGG, FLAC)
Traces uploadJSON trace files with transcripts and recording URLs
Each source creates call records that flow through the same analysis pipeline regardless of how they were ingested.

​
Labels

You can tag calls with labels to organize and filter your data. Labels can be applied during sync, upload, or after ingestion. Use them to scope monitor reports to specific subsets of calls.

​
Next steps

Set up integrations

Connect your voice AI provider to SuperBryn.

Call analysis

Understand what SuperBryn measures and audits.

Monitor reports

Generate bulk audit reports across your calls.

Custom metrics

Define your own evaluation criteria.