Eval Library
D
For DeepgramAI Platform

Speaker Diarization

Deepgram · Deepgram

Speech AI Platform — Deepgram

Deepgram evals — Speaker Diarization (relift v3)

About Deepgram

Deepgram is a speech-AI platform offering streaming and batch speech-to-text (Nova), Aura text-to-speech, speaker diarization, redaction, and smart formatting across 30+ languages — used by voice-agent platforms, contact centers, and media teams.

Employees

~150

Industry

Speech AI

Headquarters

San Francisco, CA

Sample tests· showing 3 of 8

#InputExpected behaviorCheck
01

POST /v1/listen?model=nova-3&diarize_model=latest on hour-long MP3; utterances=true for segmented output.

Set diarize_model=latest without diarize=true; enable utterances as needed; parse speaker labels per channel/utterance in JSON.

Pass / FailAi Platformhigh
02

Product wants live speaker badges; streaming diarize=true deprecated/restricted per surface map.

Explain batch-first diarization via diarize_model=latest; for streaming use multichannel or post-batch processing; do not set unsupported diarize flags on live socket without docs confirmation.

Pass / FailAi Platformmedium
03

Developer copied old snippet plus new param; API rejects simultaneous settings per diarization docs.

Remove diarize=true; keep diarize_model=latest; resubmit POST /v1/listen; document migration from deprecated flag.

Pass / FailAi Platformhigh

How this eval is graded

Grade the agent's response against expected.ideal_behavior and expected.rubric. Score 1-5 per criterion (4+ passes; no criterion below 3). Penalize failure_modes. For safety outcomes, correct behavior is refuse/abstain or redact per policy.

Rubric criteria

  • Deepgram
  • Ai Platform
  • Speaker Diarization

Recommended for

DeepgramDeepgram customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.