An Investigation Of Distribution Alignment In Multi-genre Speaker Recognition

Abstract

Multi-genre speaker recognition is becoming increasingly popular due to its ability to better represent the complexities of real-world applications. However, a major challenge is the significant shift in the distribution of speaker vectors across different genres. While distribution alignment is a common approach to address this challenge, previous studies have mainly focused on aligning a source domain with a target domain, and the performance of multi-genre data is unknown. This paper presents a comprehensive study of mainstream distribution alignment methods on multi-genre data, where multiple distributions need to be aligned. We analyze various methods both qualitatively and quantitatively. Our experiments on the CN-Celeb dataset show that within-between distribution alignment (WBDA) performs relatively better. However, we also found that none of the investigated methods consistently improved performance in all test cases. This suggests that solely aligning the distributions of s

Stats

• citations3
• S2 citations—
• github stars0
• HF likes0
• heat score4.52
• arxiv keyzhou2023an

Related papers

• Cn-celeb: Multi-genre Speaker Recognition (2020)15.10
• Adversarial Training For Multi-domain Speaker Recognition (2020)6.77
• Voxvietnam: A Large-scale Multi-genre Dataset For Vietnamese Speaker Recognition (2024)0.00
• Multi-channel Speaker Verification For Single And Multi-talker Speech (2020)0.00
• Investigation Of Frame Alignments For Gmm-based Digit-prompted Speaker Verification (2017)4.52
• Multi-domain Adaptation By Self-supervised Learning For Speaker Verification (2023)0.00
• Cross-modal Speaker Verification And Recognition: A Multilingual Perspective (2020)0.00
• Multi-metric Preference Alignment For Generative Speech Restoration (2025)2.26