Automatic activity detection configuration for Live API sessions.
Configures automatic detection of user activity (voice and text input). When enabled (the default), the server automatically detects when the user starts and stops speaking.
Fields
disabled- If true, automatic detection is disabled and client must send activity signalsstart_of_speech_sensitivity- How likely speech is to be detected at startend_of_speech_sensitivity- How likely detected speech is to endprefix_padding_ms- Duration of speech required before start-of-speech is committedsilence_duration_ms- Duration of silence required before end-of-speech is committed
Example
%AutomaticActivityDetection{
disabled: false,
start_of_speech_sensitivity: :high,
end_of_speech_sensitivity: :low,
prefix_padding_ms: 100,
silence_duration_ms: 500
}
Summary
Functions
Parses from API response.
Creates a new AutomaticActivityDetection configuration.
Converts to API format (camelCase).
Types
@type t() :: %Gemini.Types.Live.AutomaticActivityDetection{ disabled: boolean() | nil, end_of_speech_sensitivity: Gemini.Types.Live.Enums.EndSensitivity.t() | nil, prefix_padding_ms: integer() | nil, silence_duration_ms: integer() | nil, start_of_speech_sensitivity: Gemini.Types.Live.Enums.StartSensitivity.t() | nil }