Remove Silence from Audio

Pick a recording — long silent gaps are stripped in your browser, nothing is uploaded.

De-silenced audio

Remove silence from audio in your browser

Pick a recording and the tool strips dead air — leading silence, long pauses in the middle, and trailing silence — using ffmpeg's silenceremove filter, entirely in your browser. A quarter second of every removed gap is kept, so speech keeps its natural rhythm instead of sounding hard-cut.

Worked example

Tighten up a podcast interview: upload interview.mp3 and leave the defaults — Silence threshold -30 dB and Min gap 0.5 seconds. Every pause longer than half a second collapses to a 0.25 s beat, and the result interview-nosilence.mp3 is often 20–40% shorter with no lost words. If a quiet room tone is being kept as "speech", lower the threshold to -40; if breaths are being cut, raise the minimum gap to 1.

How the two knobs interact

Limits and edge cases

FAQ

Why is there still a short pause at each cut?

That's deliberate: 0.25 s of every removed gap is kept so sentences don't slam into each other. Fully gapless cuts sound robotic and make speech hard to follow.

My quiet recording lost words — what should I change?

Lower the threshold (e.g. -40 or -50 dB). Your speech is quieter than the default -30 dB cutoff, so parts of it are being classified as silence. If only word endings clip, also raise the minimum gap so brief dips aren't trimmed.

Does this work for videos too?

Not this tool — cutting silence out of a video needs the audio and video streams cut identically to stay in sync. Use the separate video-silence-cut tool for that; this one is audio-only and faster because of it.

Is my audio uploaded anywhere?

No. The page downloads an ffmpeg WebAssembly build once and then processes your file locally in the browser tab — the audio never leaves your device.

Developer & Automation Access

Run it from the terminal

Same engine as this page, headless — via the gizza CLI:

gizza tool audio-silence-remove 'url=https://example.com/input' 'threshold_db=-30' 'min_silence=0.5' 'format=mp3'

New to the CLI? Get gizza →

Open it by URL

Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:

https://gizza.ai/tools/audio-silence-remove/?url=https://example.com/input&threshold_db=-30&min_silence=0.5&format=mp3