Step by Step guide: Build a Voice Agent That Never Goes Silent using ADK

Summary

Voice agents that block on slow tool calls produce dead air, causing users to assume the call dropped and hang up. The open-source Agent Development Kit (ADK) with gemini-3.8-live addresses this with two non-blocking patterns. Streaming tools use Python async generators that yield progress updates, enabling the agent to narrate stages like connecting, scanning records, and aggregating revenue while the task runs in the background. Background execution uses single-yield generators or FunctionTool with response_scheduling to deliver results when idle, allowing users to ask unrelated questions without cancellation, and benchmark sales-report workflows show first words appearing in 1.2 seconds instead of 14.4. Scheduling modes WHEN_IDLE, INTERRUPT, and SILENT control when the model reacts to incoming tool results, ensuring the agent remains conversational throughout complex operations, with production features for cancellation and idempotency.

Takeaways

1. 1

   Switching from blocking to streaming tools in ADK reduces agent first-word latency from 14.4 seconds to 1.2 seconds in benchmark sales-report workflows.

2. 2

   ADK supports three response_scheduling behaviors: WHEN_IDLE waits for a natural pause, INTERRUPT cuts into current speech, and SILENT stores results silently in conversation history.

3. 3

   User barge-in does not cancel running streaming tools; to explicitly stop a tool, the model must call the built-in stop_streaming function.

FAQ