Saved in:
Bibliographic Details
Main Authors: Wang, Chien-Chun, Yu, En-Lun, Hung, Jeih-Weih, Huang, Shih-Chieh, Chen, Berlin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.20885
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915468356354048
author Wang, Chien-Chun
Yu, En-Lun
Hung, Jeih-Weih
Huang, Shih-Chieh
Chen, Berlin
author_facet Wang, Chien-Chun
Yu, En-Lun
Hung, Jeih-Weih
Huang, Shih-Chieh
Chen, Berlin
contents Voice activity detection (VAD) is essential for speech-driven applications, but remains far from perfect in noisy and resource-limited environments. Existing methods often lack robustness to noise, and their frame-wise classification losses are only loosely coupled with the evaluation metric of VAD. To address these challenges, we propose SincQDR-VAD, a compact and robust framework that combines a Sinc-extractor front-end with a novel quadratic disparity ranking loss. The Sinc-extractor uses learnable bandpass filters to capture noise-resistant spectral features, while the ranking loss optimizes the pairwise score order between speech and non-speech frames to improve the area under the receiver operating characteristic curve (AUROC). A series of experiments conducted on representative benchmark datasets show that our framework considerably improves both AUROC and F2-Score, while using only 69% of the parameters compared to prior arts, confirming its efficiency and practical viability.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20885
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SincQDR-VAD: A Noise-Robust Voice Activity Detection Framework Leveraging Learnable Filters and Ranking-Aware Optimization
Wang, Chien-Chun
Yu, En-Lun
Hung, Jeih-Weih
Huang, Shih-Chieh
Chen, Berlin
Sound
Voice activity detection (VAD) is essential for speech-driven applications, but remains far from perfect in noisy and resource-limited environments. Existing methods often lack robustness to noise, and their frame-wise classification losses are only loosely coupled with the evaluation metric of VAD. To address these challenges, we propose SincQDR-VAD, a compact and robust framework that combines a Sinc-extractor front-end with a novel quadratic disparity ranking loss. The Sinc-extractor uses learnable bandpass filters to capture noise-resistant spectral features, while the ranking loss optimizes the pairwise score order between speech and non-speech frames to improve the area under the receiver operating characteristic curve (AUROC). A series of experiments conducted on representative benchmark datasets show that our framework considerably improves both AUROC and F2-Score, while using only 69% of the parameters compared to prior arts, confirming its efficiency and practical viability.
title SincQDR-VAD: A Noise-Robust Voice Activity Detection Framework Leveraging Learnable Filters and Ranking-Aware Optimization
topic Sound
url https://arxiv.org/abs/2508.20885