Table of Contents

Namespace LMKit.Speech

Classes

AudioSegment

Represents a segment of audio with its recognized text, timing, confidence score, and language.

SpeechToText

Provides transcription and language-detection capabilities using an LM model with speech-to-text support.

SpeechToText.LanguageDetectionResult

Represents the result of a language detection operation, including the detected language code and the confidence score.

SpeechToText.OnNewSegmentEventArgs

Event arguments for when a new AudioSegment is produced during transcription.

SpeechToText.OnProgressEventArgs

Provides data for the OnProgress event, containing the current completion percentage of the transcription.

SpeechToText.TranscriptionResult

Holds the set of AudioSegments produced by a transcription and provides the full combined text.

VadSettings

Configuration settings for voice activity detection (VAD).

Enums

SpeechToText.SpeechToTextMode

Specifies the operating mode for the SpeechToText engine: whether to transcribe speech in the original language or to translate it into English.