Automatically find awkward pauses and filler words, fine-tune the sensitivity, preview every cut, and export a clean version — all processed locally with the Web Audio API and FFmpeg (WebAssembly). Nothing is uploaded.
Upload a file, or record straight from your microphone
MP3, WAV, MP4, WEBM, M4A — processed entirely in your browser
No. Analysis runs with the Web Audio API and the actual trimming/export runs through FFmpeg compiled to WebAssembly, both entirely inside your browser tab. Nothing is sent to a server.
The tool measures the audio's loudness in small time windows and flags stretches that stay below your chosen sensitivity threshold for longer than the minimum silence length you set.
Automatic filler-word tagging uses your browser's live speech recognition, which only works on a live microphone stream — not on a file that's already saved. Record directly in this tool to get filler words tagged automatically; uploaded files rely on silence-based trimming, and you can always mark extra cuts manually in the preview list.
Yes. Every detected cut appears in the manual review list where you can enable or disable it individually, then use "Preview With Cuts Applied" to listen before exporting anything.
Video input is exported as .webm; audio-only input is exported as .mp3. Both play natively in modern browsers.