Leone Venticinque·Aug 17Quanti modi ha l’IA per risolvere un problema? (2)Un veloce esperimento di analisi sonora
Hemang Baldha·Aug 13Building a Deep Learning-Based Voice Activity Detection Model from ScratchBuilding a Deep Learning-Based Voice Activity Detection Model from Scratch
Aditya Arya·Jul 22From Walkie-Talkie to True-Duplex: The VAD wayGitHub: https://github.com/Adityarya11
InAgora.iobyAkshay Nandwana·Jul 9What is VAD? A Practical Guide to Voice Activity Detection for Real-Time Voice AI AppsVoice AI is no longer only about converting speech to text.A response icon1A response icon1
Atharsh K·Jul 4Why Cascading Voice Agents Feel Broken, and the Architecture That Actually Fixes ItResearch notes from building a real-time voice triage agent, covering the landscape, the failure modes, and a full architecture breakdown…A response icon1A response icon1
Marcus Chen·Jul 2The demo was flawless. The first real call had three people talking.A scripted voice-agent demo works with one clean speaker. The first production call had crosstalk, and turn-taking fell apart. This is how…
AlterSquare·Jun 1Why VAD End-of-Speech Detection Is the Hardest Problem in Production Voice AgentsTL;DR: VAD decides when a user has stopped speaking — and getting it wrong kills the conversation. Models trained on clean audio fail in…
AlterSquare·May 20Voice Agent Latency: Where the 2–3 Second Delay Actually Lives in the Pipeline and How to Reduce ItTL;DR: The 2–3 second delay in voice agents is not random — it’s five sequential stages stacking up: VAD, STT, LLM, TTS, and network…
Peter Lebedev·Apr 20Pitch Detection and Auto-CalibrationThe three-metric detector caught most noise. These two additions handled the rest.
Peter Lebedev·Apr 18Energy Alone Is Not Enough — The Signal Processing Behind a Voice Activity DetectorNinety percent accuracy sounds good until you see what the failing 10% does to your downstream pipeline.
