The Perception of Phase Intercept Distortion and its Application in Data Augmentation

TL;DR

This paper investigates phase-intercept distortion, demonstrating its imperceptibility to humans and leveraging it for effective data augmentation in audio machine learning, leading to improved performance.

Contribution

It introduces phase-intercept distortion as a novel, imperceptible form of phase distortion and applies it for data augmentation in audio machine learning tasks.

Findings

Human subjects could not perceive the distortion.

Data augmentation with this distortion improved model accuracy.

The approach is effective across multiple audio tasks.

Abstract

Phase distortion refers to the alteration of the phase relationships between frequencies in a signal, which can be perceptible. In this paper, we discuss a special case of phase distortion known as phase-intercept distortion, which is created by a frequency-independent phase shift. We hypothesize that, though this form of distortion changes a signal's waveform significantly, the distortion is imperceptible. Human-subject experiment results are reported which are consistent with this hypothesis. Furthermore, we discuss how the imperceptibility of phase-intercept distortion can be useful for machine learning, specifically for data augmentation. We conducted multiple experiments using phase-intercept distortion as a novel approach to data augmentation, and obtained improved results for audio machine learning tasks.