Window Size Versus Accuracy Experiments in Voice Activity Detectors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: McKinnon, Max, Khaki, Samir, Reddy, Chandan KA, Huang, William
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911396048928768
author McKinnon, Max
Khaki, Samir
Reddy, Chandan KA
Huang, William
author_facet McKinnon, Max
Khaki, Samir
Reddy, Chandan KA
Huang, William
contents Voice activity detection (VAD) plays a vital role in enabling applications such as speech recognition. We analyze the impact of window size on the accuracy of three VAD algorithms: Silero, WebRTC, and Root Mean Square (RMS) across a set of diverse real-world digital audio streams. We additionally explore the use of hysteresis on top of each VAD output. Our results offer practical references for optimizing VAD systems. Silero significantly outperforms WebRTC and RMS, and hysteresis provides a benefit for WebRTC.
format Preprint
id arxiv_https___arxiv_org_abs_2601_17270
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Window Size Versus Accuracy Experiments in Voice Activity Detectors
McKinnon, Max
Khaki, Samir
Reddy, Chandan KA
Huang, William
Sound
Computation and Language
Audio and Speech Processing
68T10
I.2.7; H.5.5
Voice activity detection (VAD) plays a vital role in enabling applications such as speech recognition. We analyze the impact of window size on the accuracy of three VAD algorithms: Silero, WebRTC, and Root Mean Square (RMS) across a set of diverse real-world digital audio streams. We additionally explore the use of hysteresis on top of each VAD output. Our results offer practical references for optimizing VAD systems. Silero significantly outperforms WebRTC and RMS, and hysteresis provides a benefit for WebRTC.
title Window Size Versus Accuracy Experiments in Voice Activity Detectors
topic Sound
Computation and Language
Audio and Speech Processing
68T10
I.2.7; H.5.5
url https://arxiv.org/abs/2601.17270