Configuration for endpointing, which determines when the user's turn is complete.

interface EndpointingOptions {
    alpha: number;
    maxDelay: number;
    minDelay: number;
    mode: "fixed" | "dynamic";
}

Properties

alpha: number

Exponential moving average coefficient for dynamic endpointing. Higher values give more weight to history.

Default Value

0.9
maxDelay: number

Maximum time in milliseconds the agent will wait before terminating the turn. In dynamic mode this is a fixed ceiling.

Default Value

3000
minDelay: number

Minimum time in milliseconds since the last detected speech before the agent declares the user's turn complete. In dynamic mode this is the learned floor. In VAD mode this effectively behaves like max(VAD silence, minDelay); in STT mode it is applied after the STT end-of-speech signal, so it can be additive with the STT provider's endpointing delay.

Default Value

500
mode: "fixed" | "dynamic"

Endpointing mode. "fixed" uses a fixed delay, "dynamic" adjusts delay based on end-of-utterance prediction.

Default Value

"fixed"