MediaPipeTasksAudio Framework Reference

Classes

The following classes are available globally.

• AudioClassifier

  Performs classification on audio clips or continuous audio streams.

  Declaration

  Swift

  class AudioClassifier : NSObject

• AudioClassifierLiveStreamDelegate

  Protocol adopted by delegates of AudioClassifier to receive asynchronous classification results in .audioStream mode.

  Declaration

  Swift

  protocol AudioClassifierLiveStreamDelegate : NSObjectProtocol

• AudioClassifierOptions

  Options for setting up an AudioClassifier.

  Declaration

  Swift

  class AudioClassifierOptions : TaskOptions, NSCopying

• AudioClassifierResult

  Represents the classification results generated by AudioClassifier.

  Declaration

  Swift

  class AudioClassifierResult : TaskResult

• AudioData

  Wrapper class for input audio samples backed by a ring buffer.

  Declaration

  Swift

  class AudioData : NSObject

• AudioDataFormat

  Describes the channel count and sample rate of incoming audio samples.

  Declaration

  Swift

  class AudioDataFormat : NSObject

• AudioEmbedder

  Performs audio embedding extraction on audio clips or continuous audio streams.

  Declaration

  Swift

  class AudioEmbedder : NSObject

• AudioEmbedderLiveStreamDelegate

  Protocol adopted by delegates of AudioEmbedder to receive asynchronous embedding results in .audioStream mode.

  Declaration

  Swift

  protocol AudioEmbedderLiveStreamDelegate : NSObjectProtocol

• AudioEmbedderOptions

  Options for setting up an AudioEmbedder.

  Declaration

  Swift

  class AudioEmbedderOptions : TaskOptions, NSCopying

• AudioEmbedderResult

  Represents the embedding results generated by AudioEmbedder.

  Declaration

  Swift

  class AudioEmbedderResult : TaskResult

• AudioRecord

  Wrapper class that records audio samples continuously from the device microphone.

  Declaration

  Swift

  class AudioRecord : NSObject

• AudioTranscriber

  Performs automatic speech recognition (ASR) on streaming mono PCM float audio samples.

  Declaration

  Swift

  class AudioTranscriber : NSObject

• AudioTranscriberOptions

  Options for configuring a MediaPipe AudioTranscriber task.

  Declaration

  Swift

  class AudioTranscriberOptions : TaskOptions, NSCopying

• AudioTranscriberResult

  Represents the transcription results generated by AudioTranscriber.

  Declaration

  Swift

  class AudioTranscriberResult : TaskResult

• BaseOptions

  Holds the base options used for creation of any MediaPipe task, including model asset path and hardware delegate configuration.

  Declaration

  Swift

  class BaseOptions : NSObject, NSCopying

• ClassificationResult

  Represents the classification results of a model for a single timestamped chunk.

  Declaration

  Swift

  class ClassificationResult : NSObject

• Classifications

  Represents the list of predicted categories for a given classifier head.

  Declaration

  Swift

  class Classifications : NSObject

• Embedding

  Represents the embedding extracted by an embedder task head.

  Declaration

  Swift

  class Embedding : NSObject

• EmbeddingResult

  Represents the embedding results of a model.

  Declaration

  Swift

  class EmbeddingResult : NSObject

• ResultCategory

  Contains a category label, display name, score, and index.

  Declaration

  Swift

  class ResultCategory : NSObject

• TextToSpeech

  MediaPipe TextToSpeech task interface for Swift (iOS / macOS).

  Declaration

  Swift

  final class TextToSpeech : @unchecked Sendable

• TextToSpeechOptions

  Task initialization options for TextToSpeech.

  Declaration

  Swift

  public struct TextToSpeechOptions : Sendable, Equatable

• TranscriptionSegment

  A single transcribed segment emitted by AudioTranscriber.

  Declaration

  Swift

  class TranscriptionSegment : NSObject

• TtsAudioChunk

  Incremental mono PCM audio chunk emitted by TextToSpeech.synthesizeStream and TtsStream.readNext (also available as AudioChunk).

  Declaration

  Swift

  public struct TtsAudioChunk : Sendable, Equatable

• TtsAudioResult

  Self-contained mono PCM audio buffer returned by TextToSpeech.synthesize (also available as AudioResult).

  Declaration

  Swift

  public struct TtsAudioResult : Sendable, Equatable

• TtsStream

  Bidirectional / pull-based streaming synthesis handle supporting incremental LLM text pushing (pushText + finishInput) alongside readNext() and cancel() (also available as TTSStream).

  Declaration

  Swift

  public final class TtsStream : @unchecked Sendable

• TtsSynthesisOptions

  Per-utterance synthesis overrides passed to TextToSpeech (also available as SynthesisOptions).

  Declaration

  Swift

  public struct TtsSynthesisOptions : Sendable, Equatable

• TtsVoiceInfo

  Voice metadata descriptor returned by TextToSpeech.listVoices() (also available as VoiceInfo).

  Declaration

  Swift

  public struct TtsVoiceInfo : Sendable, Equatable

• TtsWordTimestamp

  Word-level timing alignment mapping an input word to audio seconds (also available as WordTimestamp).

  Declaration

  Swift

  public struct TtsWordTimestamp : Sendable, Equatable