Low-Resource Domain Adaptation for Speaker Recognition Using Cycle-GANs

TL;DR

This paper explores the use of CycleGANs for low-resource domain adaptation in speaker recognition, demonstrating improvements in challenging scenarios like reverberant to clean speech with limited target data.

Contribution

It introduces the application of CycleGAN-based unsupervised domain adaptation in low-resource settings for speaker recognition, including novel reverberant to clean adaptation.

Findings

Achieved 18.3% relative EER reduction on VOiCES dataset

Improved over state-of-the-art WPE de-reverberation

Effective even with limited target domain data

Abstract

Current speaker recognition technology provides great performance with the x-vector approach. However, performance decreases when the evaluation domain is different from the training domain, an issue usually addressed with domain adaptation approaches. Recently, unsupervised domain adaptation using cycle-consistent Generative Adversarial Netorks (CycleGAN) has received a lot of attention. CycleGAN learn mappings between features of two domains given non-parallel data. We investigate their effectiveness in low resource scenario i.e. when limited amount of target domain data is available for adaptation, a case unexplored in previous works. We experiment with two adaptation tasks: microphone to telephone and a novel reverberant to clean adaptation with the end goal of improving speaker recognition performance. Number of speakers present in source and target domains are 7000 and 191 respectively. By adding noise to the target domain during CycleGAN training, we were able to achieve better performance compared to the adaptation system whose CycleGAN was trained on a larger target data. On reverberant to clean adaptation task, our models improved EER by 18.3% relative on VOiCES dataset compared to a system trained on clean data. They also slightly improved over the state-of-the-art Weighted Prediction Error (WPE) de-reverberation algorithm.