MDA TCH PE Peter Lebedev Energy Alone Is Not Enough — The Signal Processing Behind a Voice Activity Detector Ninety percent accuracy sounds good until you see what the failing 10% does to your downstream pipeline.
SOC MDA TCH PE Peter Lebedev Taming Telephony Audio — Mu-law, PCM, and the Conversion Gauntlet Before you can detect voice activity, you have to understand what you’re looking at. What I thought was PCM was something else entirely.
MDA PE Peter Lebedev Why I Built a Custom Voice Activity Detector in Java Building real-time speech detection for telephony audio — from energy thresholds to multi-feature adaptive calibration.
MDA PE Peter Lebedev Pitch Detection and Auto-Calibration The three-metric detector caught most noise. These two additions handled the rest.
