I built a silence/dead-air remover for podcast and voiceover editing that runs 100% client-side using WebAssembly and the Web Audio API — no server upload, no account, processes in-browser regardless of file size (limited only by your device RAM).
It scans the waveform, detects segments below a configurable dB threshold, and lets you fine-tune cuts with safety padding so it doesn't clip speech. Waveform zoom goes up to 300x for precise manual adjustment where needed.
