stt

type AvailabilityChangedEvent struct {
	STT       STT
	Available bool
}

type AvailabilityChangedHandler func(AvailabilityChangedEvent)

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

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

type FallbackAdapterOptions struct {
	MaxRetryPerSTT int
	AttemptTimeout time.Duration
	RetryInterval  time.Duration
}

type FallbackAllFailedError struct {
	Count    int
	Labels   []string
	Duration time.Duration
	Err      error
}

type InputEnding interface {
	EndInput() error
}

type MetricsEmitter 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 RecognitionUsage struct {
	AudioDuration float64 `json:"audio_duration"`
	InputTokens   int     `json:"input_tokens"`
	OutputTokens  int     `json:"output_tokens"`
}

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   `json:"streaming"`
	InterimResults    bool   `json:"interim_results"`
	Diarization       bool   `json:"diarization"`
	AlignedTranscript string `json:"aligned_transcript"`
	OfflineRecognize  bool   `json:"offline_recognize"`
}

type STTError struct {
	Type        string
	Timestamp   time.Time
	Label       string
	Err         error
	Recoverable bool
}

type STTErrorHandler func(*STTError)

type STTMetricsHandler func(*telemetry.STTMetrics)

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

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

type SpeechData struct {
	Language         string         `json:"language"`
	Text             string         `json:"text"`
	StartTime        float64        `json:"start_time"`
	EndTime          float64        `json:"end_time"`
	Confidence       float64        `json:"confidence"`
	SpeakerID        string         `json:"speaker_id"`
	IsPrimarySpeaker *bool          `json:"is_primary_speaker"`
	Words            []TimedString  `json:"words"`
	SourceLanguages  []string       `json:"source_languages"`
	SourceTexts      []string       `json:"source_texts"`
	TargetLanguages  []string       `json:"target_languages"`
	TargetTexts      []string       `json:"target_texts"`
	Metadata         map[string]any `json:"metadata"`
}

type SpeechEvent struct {
	Type             SpeechEventType   `json:"type"`
	RequestID        string            `json:"request_id"`
	Alternatives     []SpeechData      `json:"alternatives"`
	RecognitionUsage *RecognitionUsage `json:"recognition_usage"`
	SpeechStartTime  *float64          `json:"speech_start_time"`
	Interrupted      bool              `json:"-"`
}

type SpeechEventType string

type SpeechStream = RecognizeStream

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

type StreamAdapterWrapper = streamAdapterWrapper

type StreamTiming interface {
	StartTimeOffset() float64
	SetStartTimeOffset(offset float64)
	StartTime() float64
	SetStartTime(startTime float64)
}

type TimedString struct {
	Text            string  `json:"text"`
	StartTime       float64 `json:"start_time"`
	EndTime         float64 `json:"end_time"`
	Confidence      float64 `json:"confidence"`
	StartTimeOffset float64 `json:"start_time_offset"`
	SpeakerID       string  `json:"speaker_id"`
}