Training "on the fly" to improve the performance of speaker recognition in noisy environments

Al-Noori, AHY, Duncan, PJ and Li, FF 2017, 'Training "on the fly" to improve the performance of speaker recognition in noisy environments' , in: Proceedings: 2017 AES International Conference on Audio Forensics , Audio Engineering Society.

[img] PDF - Published Version
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

Reliability of Speaker Recognition (SR) is crucial for critical applications, especially in adverse acoustic conditions. Ambient noises and their variations represent a significant challenge for such applications. In this paper, a new technique is proposed to address the issue of performance degradation in noisy environments. Based on the estimation of the signal to noise ratio (SNR) and profile of the ambient noise from input signals, the proposed method re-trains the enrolment model for the claim speaker to generate new noisy models that adapt to the noise profile. This technique is termed “training on the fly”. Evaluation results show notable enhancement in performance in terms of the reduction of equal error rates over a range of SNRs and different types of noise.

Item Type: Book Section
Schools: Schools > School of Computing, Science and Engineering
Journal or Publication Title: 2017 AES International Conference on Audio Forensics
Publisher: Audio Engineering Society
ISBN: 9781942220145
Related URLs:
Depositing User: Mr AHY Al-Noori
Date Deposited: 19 Jul 2017 15:40
Last Modified: 03 Oct 2017 15:26
URI: http://usir.salford.ac.uk/id/eprint/43100

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year