Voice activity detection decides, frame by frame, whether the audio contains speech or only silence and noise. That one bit drives a surprising amount of a real-time stack: it triggers discontinuous transmission to stop sending packets during silence, gates noise suppression, marks segments for transcription, and powers 'who is speaking' indicators in a conference UI. Detectors range from simple energy thresholds to neural classifiers; the engineering tension is between false negatives, which clip the start of words, and false positives, which keep transmitting noise and waste the bandwidth VAD was meant to save. Good VAD is fast, slightly biased toward keeping speech, and tuned to the noise it will face.
The book · Volume 4 of 7
Audio for Video: A Complete Guide to Sound in Video Products: From PCM and Codecs to Loudness, Accessibility, and Real-Time Voice Processing
Voice activity detection quietly drives DTX, noise gating and transcript triggers, so a bad VAD clips first syllables or wastes bandwidth on silence. Volume 4 covers VAD in detail as part of real-time voice processing, next to AEC, AGC, noise suppression and the Opus features that depend on it. Written by Nikolay Sapunov, CEO at Fora Soft.
Is your VAD clipping first words or sending noise as speech in calls?
Send us recordings of the problem and we will tune or replace your VAD and give you a tested configuration for DTX, noise suppression and transcript triggering. 250+ video products shipped since 2005.
