stt

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

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

type PrimarySpeakerDetectionOptions struct {
	FrameSizeMs            int
	RMSBufferDuration      float64
	MinRMSSamples          int
	RMSSmoothingFactor     float64
	ThresholdMultiplier    float64
	DecayToEqualTime       float64
	ThresholdMinMultiplier float64
}

type RecognizeStream interface {
	PushFrame(frame *model.AudioFrame) error
	Flush() error
	Close() error
	Next() (*SpeechEvent, error)
}

type STT interface {
	Label() string
	Capabilities() STTCapabilities
	Stream(ctx context.Context, language string) (RecognizeStream, error)
	Recognize(ctx context.Context, frames []*model.AudioFrame, language string) (*SpeechEvent, error)
}

type STTCapabilities struct {
	Streaming        bool
	InterimResults   bool
	Diarization      bool
	OfflineRecognize bool
}

type SearchStream interface {
	PushFrame(frame *model.AudioFrame) error
	Close() error
	Next() (*SpeechEvent, error)
}

type SpeechData struct {
	Language         string
	Text             string
	StartTime        float64
	EndTime          float64
	Confidence       float64
	SpeakerID        string
	IsPrimarySpeaker *bool
}

type SpeechEvent struct {
	Type         SpeechEventType
	RequestID    string
	Alternatives []SpeechData
	Interrupted  bool
}

type SpeechEventType string

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