Contents

featureExtractionTimeWindowSize

A time duration, in seconds, that determines how much audio data the feature-extraction session reads each time it samples an audio file.

Declaration

var featureExtractionTimeWindowSize: TimeInterval { get set }

Discussion

The time-window size defaults to 0.975 seconds and must be in the range [0.5, 15.0].

Feature-extraction sessions that use MLSoundClassifier.ModelParameters.FeatureExtractorType.vggish(revision:) ignore this value and always use a time-window size of 0.975 seconds.

See Also

Accessing feature extraction parameters