Time-Contrastive Learning Based DNN Bottleneck Features for Text-Dependent Speaker Verification

TL;DR

This paper introduces a novel time-contrastive learning approach for DNN bottleneck feature extraction that leverages temporal structure in speech, improving text-dependent speaker verification performance.

Contribution

It proposes a TCL-based BN feature extraction method that learns generic features from unlabeled temporal segments, outperforming traditional speaker and pass-phrase discriminant features.

Findings

TCL-BN features outperform existing BN and MFCC features in speaker verification.

The method effectively captures temporal structure for robust feature learning.

Experimental results on RedDots Challenge 2016 validate the approach's superiority.

Abstract

In this paper, we present a time-contrastive learning (TCL) based bottleneck (BN)feature extraction method for speech signals with an application to text-dependent (TD) speaker verification (SV). It is well-known that speech signals exhibit quasi-stationary behavior in and only in a short interval, and the TCL method aims to exploit this temporal structure. More specifically, it trains deep neural networks (DNNs) to discriminate temporal events obtained by uniformly segmenting speech signals, in contrast to existing DNN based BN feature extraction methods that train DNNs using labeled data to discriminate speakers or pass-phrases or phones or a combination of them. In the context of speaker verification, speech data of fixed pass-phrases are used for TCL-BN training, while the pass-phrases used for TCL-BN training are excluded from being used for SV, so that the learned features can be considered generic. The method is evaluated on the RedDots Challenge 2016 database. Experimental results show that TCL-BN is superior to the existing speaker and pass-phrase discriminant BN features and the Mel-frequency cepstral coefficient feature for text-dependent speaker verification.