speech

type Detector struct {
	// contains filtered or unexported fields
}

type DetectorConfig struct {
	// The path to the ONNX Silero VAD model file to load.
	ModelPath string
	// The sampling rate of the input audio samples. Supported values are 8000 and 16000.
	SampleRate int
	// The probability threshold above which we detect speech. A good default is 0.5.
	Threshold float32
	// The duration of silence to wait for each speech segment before separating it.
	MinSilenceDurationMs int
	// The padding to add to speech segments to avoid aggressive cutting.
	SpeechPadMs int
	// The loglevel for the onnx environment, by default it is set to LogLevelWarn.
	LogLevel LogLevel
}

type LogLevel int

type Segment struct {
	// The relative timestamp in seconds of when a speech segment begins.
	SpeechStartAt float64
	// The relative timestamp in seconds of when a speech segment ends.
	SpeechEndAt float64
}