Remove Silence from Audio

Cut silent gaps and long pauses out of podcasts, lectures and voice recordings automatically. Set the silence level, minimum pause length and padding, see every cut on the waveform, then download MP3 or WAV — free, no signup, no upload.

Input

Drop an audio file here

or choose one from your device

MP3, WAV, OGG, AAC, M4A, FLAC or WebM · Up to 500 MB

How to Remove Silence from Audio

  1. Drop your audio file (MP3, WAV, OGG, AAC, M4A, FLAC or WebM, up to 500MB) onto the input area, or choose it with the button
  2. Look at the waveform: the parts shown in red are the silences that will be cut
  3. Adjust the silence level, the shortest pause to cut and the padding until only the gaps you want are red
  4. Pick MP3 or WAV, press Remove Silence, listen to the result, then download it

Tips

  • For speech, start at −40 dB with a 500 ms shortest pause and 100 ms of padding
  • If a noisy room hides the pauses, raise the silence level a few dB above the quiet level shown under the slider
  • If sentences feel rushed, add padding or raise the shortest pause so short breaths stay
  • Choose WAV if you will keep editing the audio, MP3 to share or upload it

How Automatic Silence Removal Works

What this tool does

Recordings of talks, podcasts, interviews and voice memos are full of dead air: the seconds before someone starts speaking, the pause while a guest thinks, the gap after a slide change. This tool finds those quiet stretches and joins the parts that remain, so a 40-minute lecture with long pauses can become a tighter 32-minute file without editing it by hand.

Everything happens inside your browser. Your audio is decoded and processed on your own device; nothing is sent to a server, and there is no account, watermark or limit on how many files you clean up.

How silence is detected

The audio is split into 20-millisecond frames and the loudness (RMS level) of each frame is measured. A frame counts as silent when it is quieter than the silence level you set — for stereo files the louder channel decides, so a sound on only one side is never cut. A run of silent frames becomes a cut only if it lasts at least as long as the shortest pause you chose.

Padding then hands back a little silence on each side of every cut, so the last syllable before a pause and the breath after it stay natural. Where two pieces are joined, an 8-millisecond crossfade smooths the edit so you don't hear a click.

Example: tightening a recorded lecture

A 45-minute classroom recording has a steady hum from the projector, so its quiet parts sit around −52 dB. With the default −40 dB level the pauses between explanations are found easily; the summary shows 210 silent parts and a result about 7 minutes shorter. Raising the shortest pause to 1,200 ms keeps the speaker's natural rhythm and only removes the long waits while they write on the board.

After pressing Remove Silence, the result plays right in the output panel. If it sounds rushed, add 50 ms of padding and run it again — the waveform updates as you move each slider, so you can see the effect before processing.

Notes and limits

The tool removes pauses, not filler words: an "um" or a cough is louder than silence and stays. Very noisy recordings may need a higher silence level, and music with quiet passages can lose intentional rests, so check the red parts before you export. Files up to 500MB are accepted, but long WAV files use a lot of memory; if your browser runs out, split the recording first with the audio trimmer. Tags such as title and cover art are not copied to the new file, and the sample rate follows your device's audio setting (usually 44.1 or 48 kHz).

Frequently Asked Questions

No. The file is decoded and cut on your own device, and the page makes no upload request with your audio — you can confirm this in your browser's network panel while it runs. That makes it safe for interviews, meetings and other private recordings.