Self-Supervised Open-Set Speaker Recognition with Laguerre-Voronoi Descriptors

• Published in: Sensors (Basel, Switzerland) Vol. 24; no. 6; p. 1996
• Main Authors: Ohi, Abu Quwsar, Gavrilova, Marina L
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 21.03.2024; MDPI
• Subjects: behavioral biometric; Biometry; Computational linguistics; deep neural network; Investigations; Laguerre–Voronoi diagram; Language processing; Machine learning; Natural language interfaces; Neural networks; open-set speaker recognition; representation learning; self-supervised learning; Speech; New Jersey; representation learning; Laguerre–Voronoi diagram; behavioral biometric; open-set speaker recognition; self-supervised learning; deep neural network; smart sensors
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s24061996

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.