|---| | [ View source on GitHub](https://github.com/google-ai-edge/mediapipe/blob/master/mediapipe/tasks/python/audio/__init__.py) | MediaPipe Tasks Audio API. ## Classes [`class AudioClassifier`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/AudioClassifier): Class that performs audio classification on audio data. [`class AudioClassifierOptions`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/AudioClassifierOptions): Options for the audio classifier task. [`class AudioClassifierResult`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/AudioClassifierResult): Contains the classification results of a model. [`class AudioTranscriber`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/AudioTranscriber): Performs streaming and unary speech-to-text transcription. [`class AudioTranscriberOptions`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/AudioTranscriberOptions): Options for MediaPipe AudioTranscriber task. [`class AudioTranscriberResult`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/AudioTranscriberResult): Transcription result returned by `AudioTranscriber`. [`class ModelArchitecture`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/ModelArchitecture): Enum where members are also (and must be) ints [`class RunningMode`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/RunningMode): MediaPipe audio task running mode. [`class TextToSpeech`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/TextToSpeech): On-device speech synthesis via MediaPipe CalculatorGraph. [`class TextToSpeechOptions`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/TextToSpeechOptions): Options for MediaPipe TextToSpeech task. [`class TranscriptionSegment`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/TranscriptionSegment): A single transcribed segment with start and end timestamps. [`class TtsAudioResult`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/TtsAudioResult): Synthesized audio chunk or full clip returned by `TextToSpeech`. [`class TtsStream`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/TtsStream): Bidirectional streaming synthesis session supporting `push_text()`. [`class VoiceInfo`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/VoiceInfo): Metadata for a speaker voice available in a TTS model bundle. [`class WordTimestamp`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/WordTimestamp): Word-level timing alignment mapping an input word to audio seconds. ## Functions [`compute_word_timestamps(...)`](https://developers.google.com/edge/api/mediapipe/python/mp/tasks/audio/compute_word_timestamps): Computes prosody- and waveform-aligned word timestamps via the C API.