Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification

The performance of speaker-related systems usually degrades heavily in practical applications largely due to the background noise. To improve the robustness of such systems in unknown noisy environments, this paper proposes a simple pre-processing method called Noise Invariant Frame Selection (NIFS)...

Full description

Bibliographic Details
Main Authors: Song, Siyang, Shuimei, Zhang, Schuller, Björn, Shen, Linlin, Valstar, Michel F.
Format: Conference or Workshop Item
Published: 2018
Online Access:https://eprints.nottingham.ac.uk/51473/
_version_ 1848798504382627840
author Song, Siyang
Shuimei, Zhang
Schuller, Björn
Shen, Linlin
Valstar, Michel F.
author_facet Song, Siyang
Shuimei, Zhang
Schuller, Björn
Shen, Linlin
Valstar, Michel F.
author_sort Song, Siyang
building Nottingham Research Data Repository
collection Online Access
description The performance of speaker-related systems usually degrades heavily in practical applications largely due to the background noise. To improve the robustness of such systems in unknown noisy environments, this paper proposes a simple pre-processing method called Noise Invariant Frame Selection (NIFS). Based on several noisy constraints, it selects noise invariant frames from utterances to represent speakers. Experiments conducted on the TIMIT database showed that the NIFS can significantly improve the performance of Vector Quantization (VQ), Gaussian Mixture Model-Universal Background Model (GMM-UBM) and i-vector-based speaker verification systems in different unknown noisy environments with different SNRs, in comparison to their baselines. Meanwhile, the proposed NIFS-based speaker systems has achieves similar performance when we change the constraints (hyper-parameters) or features, which indicates that it is easy to reproduce. Since NIFS is designed as a general algorithm, it could be further applied to other similar tasks.
first_indexed 2025-11-14T20:20:49Z
format Conference or Workshop Item
id nottingham-51473
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:20:49Z
publishDate 2018
recordtype eprints
repository_type Digital Repository
spelling nottingham-514732020-05-04T19:45:39Z https://eprints.nottingham.ac.uk/51473/ Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification Song, Siyang Shuimei, Zhang Schuller, Björn Shen, Linlin Valstar, Michel F. The performance of speaker-related systems usually degrades heavily in practical applications largely due to the background noise. To improve the robustness of such systems in unknown noisy environments, this paper proposes a simple pre-processing method called Noise Invariant Frame Selection (NIFS). Based on several noisy constraints, it selects noise invariant frames from utterances to represent speakers. Experiments conducted on the TIMIT database showed that the NIFS can significantly improve the performance of Vector Quantization (VQ), Gaussian Mixture Model-Universal Background Model (GMM-UBM) and i-vector-based speaker verification systems in different unknown noisy environments with different SNRs, in comparison to their baselines. Meanwhile, the proposed NIFS-based speaker systems has achieves similar performance when we change the constraints (hyper-parameters) or features, which indicates that it is easy to reproduce. Since NIFS is designed as a general algorithm, it could be further applied to other similar tasks. 2018-07-08 Conference or Workshop Item PeerReviewed Song, Siyang, Shuimei, Zhang, Schuller, Björn, Shen, Linlin and Valstar, Michel F. (2018) Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification. In: International Joint Conference on Neural Networks 2018, 8-13 July 2018, Rio de Janeiro, Brazil. (In Press)
spellingShingle Song, Siyang
Shuimei, Zhang
Schuller, Björn
Shen, Linlin
Valstar, Michel F.
Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
title Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
title_full Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
title_fullStr Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
title_full_unstemmed Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
title_short Noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
title_sort noise invariant frame selection: a simple method to address the background noise problem for text-independent speaker verification
url https://eprints.nottingham.ac.uk/51473/