Harmonic-aligned Frame Mask Based on Non-stationary Gabor Transform with Application to Content-dependent Speaker Comparison

TL;DR

This paper introduces a harmonic-aligned frame mask based on non-stationary Gabor transform for speech signals, enhancing speaker comparison by capturing harmonic features with improved alignment.

Contribution

It extends frame mask techniques to speech signals using NSGT with pitch-dependent resolution, enabling better harmonic alignment for speaker identification.

Findings

Frame masks effectively differentiate speaker identities.

Deep neural networks validate the mask's ability to represent speaker features.

Potential for speaker identification with limited data.

Abstract

We propose harmonic-aligned frame mask for speech signals using non-stationary Gabor transform (NSGT). A frame mask operates on the transfer coefficients of a signal and consequently converts the signal into a counterpart signal. It depicts the difference between the two signals. In preceding studies, frame masks based on regular Gabor transform were applied to single-note instrumental sound analysis. This study extends the frame mask approach to speech signals. For voiced speech, the fundamental frequency is usually changing consecutively over time. We employ NSGT with pitch-dependent and therefore time-varying frequency resolution to attain harmonic alignment in the transform domain and hence yield harmonic-aligned frame masks for speech signals. We propose to apply the harmonic-aligned frame mask to content-dependent speaker comparison. Frame masks, computed from voiced signals of a same vowel but from different speakers, were utilized as similarity measures to compare and distinguish the speaker identities (SID). Results obtained with deep neural networks demonstrate that the proposed frame mask is valid in representing speaker characteristics and shows a potential for SID applications in limited data scenarios.