ElevenLabs integration
Turn ElevenLabs conversation records into quality evidence.
Import real agent conversations, evaluate the behaviors that matter, and move from a dashboard trend to the transcript, available audio, and rationale behind it.
What the integration brings into review
Agent discovery
Find available ElevenLabs agents before selecting conversations to import.
Conversation evidence
Review normalized transcripts, provider metadata, and available media in one workspace.
Evaluation results
Run evaluator-backed scores and retain the rationale alongside each conversation.
Look beyond provider activity
Call duration and conversation volume are useful operational context. VaaniEval adds reviewable quality signals such as task completion, intent understanding, and required-information capture, so teams can inspect whether a call achieved the right outcome.
VaaniEval can be self-hosted under AGPL-3.0. Operators control provider permissions, evaluator configuration, retention, and access policies.
Review the AGPL-3.0 termsIntegration support
ElevenLabs, Vapi, and Bolna are supported today.
VaaniEval supports conversation imports from ElevenLabs, Vapi, and Bolna today. More integrations are coming soon.
Please do not include credentials or private call data in an integration request.
Own your benchmark
Choose voice models using evidence from your own calls.
See how a private benchmark runs inside your environment, on your production scenarios, across the models you are considering.
30 minutes. Bring your model-selection question.