AudioTranscribeWhisperVadConfiguration

Configure VAD parameters to use a VAD model integrated in whisper-cpp. These parameters are directly from the whisper VAD configuration

Field

Type

Repeated

Description

model

string

The GGML VAD model file path (required)

threshold

OptionalFloat

min_speech_duration_ms

OptionalInt

Min duration for a valid speech segment.

min_silence_duration_ms

OptionalInt

Min silence duration to consider speech as ended.

max_speech_duration_s

OptionalFloat

Max duration of a speech segment before forcing a new segment.

speech_pad_ms

OptionalInt

Padding added before and after speech segments.

samples_overlap

OptionalFloat

Overlap in seconds when copying audio samples from speech segment.

Member of