Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors
Speaker recognition is a challenging problem in behavioral biometrics that has been rigorously investigated over the last decade. Although numerous supervised closed-set systems inherit the power of deep neural networks, limited studies have been made on open-set speaker recognition. This paper prop...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-03-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/24/6/1996 |
_version_ | 1797239343055634432 |
---|---|
author | Abu Quwsar Ohi Marina L. Gavrilova |
author_facet | Abu Quwsar Ohi Marina L. Gavrilova |
author_sort | Abu Quwsar Ohi |
collection | DOAJ |
description | Speaker recognition is a challenging problem in behavioral biometrics that has been rigorously investigated over the last decade. Although numerous supervised closed-set systems inherit the power of deep neural networks, limited studies have been made on open-set speaker recognition. This paper proposes a self-supervised open-set speaker recognition that leverages the geometric properties of speaker distribution for accurate and robust speaker verification. The proposed framework consists of a deep neural network incorporating a wider viewpoint of temporal speech features and Laguerre–Voronoi diagram-based speech feature extraction. The deep neural network is trained with a specialized clustering criterion that only requires positive pairs during training. The experiments validated that the proposed system outperformed current state-of-the-art methods in open-set speaker recognition and cluster representation. |
first_indexed | 2024-04-24T17:50:01Z |
format | Article |
id | doaj.art-ecdaeae01d534bb388350a8f29628fae |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-24T17:50:01Z |
publishDate | 2024-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-ecdaeae01d534bb388350a8f29628fae2024-03-27T14:04:20ZengMDPI AGSensors1424-82202024-03-01246199610.3390/s24061996Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi DescriptorsAbu Quwsar Ohi0Marina L. Gavrilova1Department of Computer Science, University of Calgary, Calgary, AB T2N1N4, CanadaDepartment of Computer Science, University of Calgary, Calgary, AB T2N1N4, CanadaSpeaker recognition is a challenging problem in behavioral biometrics that has been rigorously investigated over the last decade. Although numerous supervised closed-set systems inherit the power of deep neural networks, limited studies have been made on open-set speaker recognition. This paper proposes a self-supervised open-set speaker recognition that leverages the geometric properties of speaker distribution for accurate and robust speaker verification. The proposed framework consists of a deep neural network incorporating a wider viewpoint of temporal speech features and Laguerre–Voronoi diagram-based speech feature extraction. The deep neural network is trained with a specialized clustering criterion that only requires positive pairs during training. The experiments validated that the proposed system outperformed current state-of-the-art methods in open-set speaker recognition and cluster representation.https://www.mdpi.com/1424-8220/24/6/1996representation learningself-supervised learningdeep neural networkLaguerre–Voronoi diagramopen-set speaker recognitionbehavioral biometric |
spellingShingle | Abu Quwsar Ohi Marina L. Gavrilova Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors Sensors representation learning self-supervised learning deep neural network Laguerre–Voronoi diagram open-set speaker recognition behavioral biometric |
title | Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors |
title_full | Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors |
title_fullStr | Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors |
title_full_unstemmed | Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors |
title_short | Self-Supervised Open-Set Speaker Recognition with Laguerre–Voronoi Descriptors |
title_sort | self supervised open set speaker recognition with laguerre voronoi descriptors |
topic | representation learning self-supervised learning deep neural network Laguerre–Voronoi diagram open-set speaker recognition behavioral biometric |
url | https://www.mdpi.com/1424-8220/24/6/1996 |
work_keys_str_mv | AT abuquwsarohi selfsupervisedopensetspeakerrecognitionwithlaguerrevoronoidescriptors AT marinalgavrilova selfsupervisedopensetspeakerrecognitionwithlaguerrevoronoidescriptors |