Smart Formatting Profanity
Deepgram · Deepgram
Speech AI Platform — Deepgram
Deepgram evals — Smart Formatting & Profanity Filter (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.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | POST /v1/listen?model=nova-3&smart_format=true on investor relations audio mentioning revenue figures. | Enable smart_format=true on batch or streaming listen; verify currency normalization in transcript text; do not post-process with regex that breaks redaction tags. | Pass / FailAi Platformmedium |
| 02 | Caller says 'five five five one two three four five six seven'; smart_format=true on streaming listen. | Enable smart_format=true; validate digit grouping in transcript; if also using redact=pii, expect phone entities redacted per policy. | Pass / FailAi Platformhigh |
| 03 | Caller speaks card number; smart_format normalizes digits while redact=numbers should mask PAN segments. | Set both smart_format=true and redact=numbers; verify numerals formatted then redacted; never persist full PAN in logs. | Pass / FailAi Platformcritical |
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
- Smart Formatting Profanity
Recommended for
Works with
Related evals
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.