Eres2netv2: Boosting Short-duration Speaker Verification Performance With Computational Efficiency

Abstract

Speaker verification systems experience significant performance degradation when tasked with short-duration trial recordings. To address this challenge, a multi-scale feature fusion approach has been proposed to effectively capture speaker characteristics from short utterances. Constrained by the model's size, a robust backbone Enhanced Res2Net (ERes2Net) combining global and local feature fusion demonstrates sub-optimal performance in short-duration speaker verification. To further improve the short-duration feature extraction capability of ERes2Net, we expand the channel dimension within each stage. However, this modification also increases the number of model parameters and computational complexity. To alleviate this problem, we propose an improved ERes2NetV2 by pruning redundant structures, ultimately reducing both the model parameters and its computational cost. A range of experiments conducted on the VoxCeleb datasets exhibits the superiority of ERes2NetV2, which achieves EER of

Stats

• citations17
• S2 citations—
• github stars0
• HF likes0
• heat score9.41
• arxiv keychen2024eres2netv2

Related papers

• An Enhanced Res2net With Local And Global Feature Fusion For Speaker Verification (2023)19.74
• ECAPA2: A Hybrid Neural Network Architecture And Training Strategy For Robust Speaker Embeddings (2024)0.00
• Unified Hypersphere Embedding For Speaker Recognition (2018)0.00
• Leveraging ASR Pretrained Conformers For Speaker Verification Through Transfer Learning And Knowledge Distillation (2023)10.74
• Rawnext: Speaker Verification System For Variable-duration Utterances With Deep Layer Aggregation And Extended Dynamic Scaling Policies (2021)12.24
• Short-segment Speaker Verification With Pre-trained Models And Multi-resolution Encoder (2025)0.00
• Neural Scoring: A Refreshed End-to-end Approach For Speaker Recognition In Complex Conditions (2024)0.00
• A Comparative Re-assessment Of Feature Extractors For Deep Speaker Embeddings (2020)8.09