On a time-frequency blurring operator with applications in data augmentation

TL;DR

This paper introduces a novel time-frequency blurring operator for data augmentation in audio signal classification, demonstrating its effectiveness in improving model performance, especially with limited data, through analysis and neural network experiments.

Contribution

The paper proposes a new time-frequency blurring operator and analyzes its mathematical properties, applying it to enhance data augmentation for audio classification tasks.

Findings

The operator improves classification accuracy in data-scarce scenarios.

Spectrogram augmentation with the operator outperforms traditional methods.

Theoretical analysis confirms the operator's boundedness and positivity.

Abstract

Inspired by the success of recent data augmentation methods for signals which act on time-frequency representations, we introduce an operator which convolves the short-time Fourier transform of a signal with a specified kernel. Analytical properties including boundedness, compactness and positivity are investigated from the perspective of time-frequency analysis. A convolutional neural network and a vision transformer are trained to classify audio signals using spectrograms with different augmentation setups, including the above mentioned time-frequency blurring operator, with results indicating that the operator can significantly improve test performance, especially in the data-starved regime.