Contents

predictions(from:overlapFactor:predictionTimeWindowSize:)

Generates predictions that use an overlap factor and time window size for an array of audio files.

Declaration

func predictions(from audioFiles: [URL], overlapFactor: Double, predictionTimeWindowSize: TimeInterval) throws -> [String]

Parameters

  • audioFiles:

    An array of audio-file URLs you want the sound classifier to categorize.

  • overlapFactor:

    The amount of overlap between successive analysis windows when the model analyzes a block of audio data.

  • predictionTimeWindowSize:

    The duration of the audio buffer the method sends to the model for each prediction.

Return Value

An array of prediction labels for the audio files.

See Also

Testing a sound classifier