Eval directory
Evals for Deepgram
14 evaluation packs covering adversarial robustness, safety gates, workflow quality, and operator-level checks for Deepgram AI products.
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.
Available eval packs for Deepgram
14 packs ready to run.
Auth Keys Rate Limits
Deepgram evals — Auth, Keys & Rate Limits (relift v3)
Batch Stt Async Callbacks
Transcription AccuracyDeepgram evals — Batch STT & Async Callbacks (relift v3)
Model Selection Language Detection
LanguageDeepgram evals — Model Selection & Language Detection (relift v3)
Pii Phi Redaction
PII LeakageDeepgram evals — PII/PHI Redaction (relift v3)
Smart Formatting Profanity
Deepgram evals — Smart Formatting & Profanity Filter (relift v3)
Speaker Diarization
Deepgram evals — Speaker Diarization (relift v3)
Aura Tts
Deepgram evals — Aura TTS (relift v3 InfraRed)
Auth Rate Limits And Concurrency
Deepgram evals — Auth, Rate Limits & Concurrency (relift v3 InfraRed)
Batch Stt Rest
Transcription AccuracyDeepgram evals — Batch STT (REST) (relift v3 InfraRed)
Safety Pii Redaction And Governance
PII LeakageDeepgram evals — Safety, PII Redaction & Governance (relift v3 InfraRed)
Speaker Diarization And Channels
Deepgram evals — Speaker Diarization & Channels (relift v3 InfraRed)
Streaming Stt Websocket
Transcription AccuracyDeepgram evals — Streaming STT (WebSocket) (relift v3 InfraRed)
Transcript Features
Deepgram evals — Transcript Features (relift v3 InfraRed)
Voice Agent Api
Deepgram evals — Voice Agent API (relift v3 InfraRed)
Why eval Deepgram AI
Deepgram's AI features ship behind brand promises about accuracy, safety, and reliability. Buyers and integrators need to know those promises hold up under adversarial prompts, edge-case workflows, and the long tail of real customer inputs — not just the demo path.
The Corsac eval library for Deepgram measures four dimensions teams care about most when deploying ai platform agents:
- Adversarial robustness — does the agent resist prompt injection, jailbreaks, and social-engineering attempts?
- Workflow quality— does it complete the task buyers were shown in the demo, on inputs that don't look like the demo?
- Safety gates — does it escalate or refuse when it should, and only then?
- Operator quality — does it preserve analyst trust by surfacing the right context at the right time?
Every eval pack above is hand-authored against Deepgram's public product surface and runnable in Corsac with your own data.