Why Do This in Your Browser?
Every podcast recording starts with a few seconds of silence while the host hits record and settles in. Every one ends with a similar tail. In a clean session those bookends are obvious; in a multi-segment edit they multiply and the manual trim work compounds. And inside the recording itself, awkward pauses, breaths, and dead-air gaps each cost a second or two — across a 30-minute episode that's minutes of wasted attention.
The fix is automated silence detection: an algorithm scans the waveform, identifies regions where amplitude stays below a threshold for a minimum duration, and removes (or shortens) those regions. FFmpeg's `silenceremove` filter does exactly this, and it runs in the wasm build with no extra dependencies. This tool exposes the two parameters that matter — threshold (how quiet counts as silence) and minimum duration (how long the gap has to be) — and applies the filter in two modes: end-trim only, or aggressive mid-clip gap compression.
How It Works
Drop an audio file. Pick a mode: 'Trim leading and trailing' (the conservative default — removes silence at the start and end only) or 'Compress mid-clip gaps' (the aggressive mode — also shortens long silent regions inside the file). Adjust the threshold in dBFS — -40dB is a good default for podcasts, -50dB for quieter content. Set the minimum silence duration that should be removed — 500ms is standard for natural-sounding edits.
FFmpeg applies the `silenceremove` filter with parameters mapped from the UI: `stop_periods=-1:stop_duration=0.5:stop_threshold=-40dB` for trim-only mode, or `start_periods=1:start_duration=0.5:start_threshold=-40dB:stop_periods=-1:stop_duration=0.5:stop_threshold=-40dB:detection=peak` for full mid-clip removal. The output shows the before-and-after duration so you can see exactly how much was cut.
Tip: For manual frame-accurate trimming rather than auto-detection, use the Audio Trimmer. After cleaning silence, run the result through the Audio Normalizer for distribution-ready loudness.
Common Use Cases
How We Compare
Honest read on free, paid, and self-hosted options for this kind of job: