Voice Agent Api
Deepgram Speech AI Platform · Deepgram
Speech AI Platform — Deepgram (Nova STT, Aura TTS, Voice Agent)
Evaluates Deepgram's Voice Agent API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Speech AI Platform eval coverage.
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 | Operator opens wss://agent.deepgram.com/agent (or documented endpoint) and sends a SettingsConfiguration message defining Aura voice, Nova STT, and an LLM brain. [REQUIRES-VERIFICATION on current endpoint host/path.] | Send a single configuration message at session start with audio (encoding, sample_rate), agent.listen (STT model), agent.think (LLM provider/model, system prompt), agent.speak (Aura voice). Validate the schema against current docs — fields shift across versions. | Pass / FailAi Platformcritical |
| 02 | User starts speaking while the agent is mid-utterance. The voice agent pipeline must stop TTS and process user audio. | Server emits a turn-event (e.g., UserStartedSpeaking) when STT detects speech onset. Stop playing buffered TTS audio immediately on receipt and flush any local audio queue. Send tts.cancel or close the TTS sub-stream per the agent contract. | Pass / FailAi Platformcritical |
| 03 | Mid-call, the LLM brain emits a function call (e.g., lookup_account). | Receive a FunctionCallRequest message with name and arguments. Execute server-side, then send a FunctionCallResponse referencing the same call id (name). Keep latency under the agent's turn budget; long-running tools should stream interim acknowledgments via Aura ('Looking that up for you'). | Pass / FailAi Platformcritical |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Deepgram
- Ai Platform
- Voice Agent Api
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Voice Agent Api eval for Deepgram Deepgram Speech AI Platform test?+
Evaluates Deepgram's Voice Agent API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Speech AI Platform eval coverage.
How is the Voice Agent Api eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Voice Agent Api pack for Deepgram Deepgram Speech AI Platform contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Voice Agent Api pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.