End-of-Turn Detection Parameters
End-of-Turn Detection Parameters
Configure Flux’s end-of-turn detection behavior with eot_threshold, eager_eot_threshold, and eot_timeout_ms.
End-of-Turn Detection Parameters
Configure Flux’s end-of-turn detection behavior with eot_threshold, eager_eot_threshold, and eot_timeout_ms.
Flux provides configurable parameters that control end-of-turn detection and language biasing, allowing you to optimize your voice agent’s conversational flow for your specific use case.
Flux’s behavior is controlled by the following key parameters:
eot_thresholdConfidence threshold required to trigger an EndOfTurn event, signaling that the user has finished speaking.
Valid Values: 0.5 to 0.9
Default: 0.7
Type: Float (passed as string in URL or SDK)
Behavior:
0.8 - 0.9) = Higher certainty required before ending a turn, fewer false positives, slightly increased latency0.5 - 0.7) = Lower certainty required before ending a turn, faster responses, more false positivesExample:
eager_eot_thresholdConfidence threshold for triggering EagerEndOfTurn events, enabling early LLM response generation.
Valid Values: 0.3 to 0.9
Default: Not set (eager mode disabled)
Type: Float (passed as string in URL or SDK)
Behavior:
EagerEndOfTurn and TurnResumed events0.3 - 0.5) = Earlier triggers, lower latency, more false starts0.6 - 0.8) = More conservative, fewer cancellations, less latency benefitTrade-offs:
EagerEndOfTurn speculative generation and TurnResumed cancellationsExample:
Important: The transcript in EagerEndOfTurn will exactly match the transcript in the subsequent EndOfTurn event (if no TurnResumed occurs). This guarantees consistency for caching strategies.
eot_timeout_msMaximum silence duration before forcing an EndOfTurn, regardless of confidence.
Valid Values: 500 to 10000 (milliseconds)
Default: 5000 (5 seconds)
Type: Integer (passed as string in URL or SDK)
Behavior:
EndOfTurn after specified silence duration, even if confidence is below eot_threshold7000 - 10000) for users with frequent pauses3000 - 4000) for rapid-response environmentsExample:
eager_eot_threshold must be less than or equal to eot_threshold (if both are set)eager_eot_threshold > eot_threshold will result in an errorBest for: Basic conversational agents, demos, getting started
Best for: High-volume customer service, fast-paced Q&A, responsiveness over accuracy
Best for: Medical/legal transcription, critical documentation, formal settings
Best for: RAG systems, tool-calling agents, multi-step reasoning workflows
Best for: Multilingual call centers, global voice agents, code-switching scenarios
See the Language Prompting guide for full details on language hint usage.