How to Evaluate Live & Voice Agents in ADK

Summary

ADK now supports native evaluation of live, voice-based agents, addressing regressions in tool use, context continuity, turn-taking, timing, and recovery that may not appear in demos. The walkthrough builds a graph-based workflow of three single-purpose agents running gemini-live-2.5-flash-native-audio, with a mocked date-of-birth tool between stages and session state carried across the open audio stream. It shows how to author scenario-based or fixed conversations, synthesize simulated user turns with Gemini TTS, configure live mode and rubric-based multi-turn trajectory judges, and run evaluations from the CLI or AgentEvaluator. ADK Web then reconstructs the audio as transcripts with playable clips, while configurable voices, languages, invocation limits, thresholds, and per-turn metrics support repeatable debugging and CI/CD regression checks.

Takeaways

1. 1

   A live workflow can sequence three ordinary agents while preserving the open audio stream, session state, and conversation history across handoffs.

2. 2

   Eval cases can use an improvising simulated user with a persona and conversation plan, or provide fixed user turns verbatim; max_allowed_invocations bounds dynamic scenarios.

3. 3

   live_model_config enables live mode, while model controls simulated-user turn taking and audio_model synthesizes speech; voices and language settings can vary audio conditions.