UtilibyWritten by the Utiliby editorial team · Last updated October 13, 2026 · Rules and formulas reviewed October 2026
What this free silence remover does
Silence is where recordings waste your time. A lecture has long pauses while the speaker writes on a board, an interview has dead air between questions, a voice memo starts and ends with seconds of nothing, and a screen recording has stretches where nobody speaks. A silence remover finds the quiet stretches and cuts them so that the audio is tighter and shorter. Doing it by hand in an editor takes patience. This free online tool finds the gaps automatically, shows them on the waveform, and lets you listen before you save.
You can remove silences completely, shorten them to a pause of your choice, or just trim the start and the end. You control the threshold, the shortest gap to cut and the amount of quiet to keep around speech. It works in your browser, so your file is not uploaded.
How to remove silence from audio in four steps
- Load your file. Drop an MP3, WAV, M4A or other audio file. The waveform and the red cut areas appear.
- Set the threshold. Anything quieter than this level counts as silence. Try Auto threshold first.
- Choose what to cut. Set the shortest silence and the padding, and pick Remove, Shorten or Trim.
- Listen and save. Press Play result. If it sounds right, choose MP3 or WAV and download.
Choosing the threshold
The threshold is measured in decibels relative to full scale (dBFS). Zero is the loudest possible signal, and quieter sounds have negative values. Speech in a good recording sits around minus 30 to minus 12 dB, and room noise is often below minus 50 dB.
| Recording | Starting threshold |
|---|---|
| Quiet studio, clean voice | -50 to -60 dB |
| Home office, light room noise | -40 to -45 dB |
| Noisy room, fans or traffic | -30 to -35 dB |
| Lecture hall or phone call | -35 to -40 dB |
If the threshold is too high, soft words and the ends of sentences are cut. If it is too low, the tool finds nothing because noise stays above it. The Auto button measures the quiet floor of your recording and sets the threshold a few decibels above it.
Remove, shorten or trim
- Remove silences: cuts each long silence down to a short pause, set by the padding. Good for podcasts and summaries where you want a quick pace.
- Shorten silences: cuts every long silence to a chosen length such as 300 ms. This keeps the rhythm of conversation but removes dead air.
- Trim start and end: removes only the silence before the first sound and after the last sound. All pauses inside stay.
For dialogue, shortening usually sounds more natural than removing, because people need small pauses to follow who is speaking and to take in what was said.
Minimum length and padding
The shortest silence setting decides which gaps count. At 500 ms, only pauses of half a second or more are touched, so the normal gaps between words stay. At 200 ms, even short pauses are cut and the result sounds fast and clipped. The padding is the quiet kept next to speech when a gap is removed. About 100 ms keeps the start and end of words clean.
Together, these two settings control the feel. A podcast edit might use minus 40 dB, 400 ms and 120 ms of padding. A lecture summary might use minus 40 dB, 600 ms and 100 ms. A tight highlight reel might use 250 ms and 60 ms.
Keeping speech natural
- Preview the result from start to finish, not just the first minute.
- Do not cut every pause. Pauses before important points give listeners time to think.
- Use Shorten for interviews with more than one speaker.
- Keep at least 80 ms of padding so consonants at the start and end of words are not clipped.
- Listen for breaths. Cutting them completely can sound robotic, so a little padding helps.
- If the audio has music or background noise under the voice, silence detection will not work, so use a noise reduction tool first.
Common uses
- Podcasts and interviews: tighten the pace before publishing.
- Lectures and webinars: shorten long recordings for review.
- Voice memos and dictation: remove dead air before transcribing, which also lowers the cost of paid transcription by the minute.
- Audiobooks and narration: trim the pauses between takes.
- Language lessons and recordings: remove silent stretches.
- Meeting recordings: skip the time when no one was talking.
Troubleshooting
- Nothing is selected: raise the threshold or click Auto threshold. Constant noise under the speech stays above a low threshold.
- Words are clipped: lower the threshold, increase the padding or lengthen the minimum silence.
- It sounds too fast: switch from Remove to Shorten, or raise the shortest silence.
- Everything is red: the threshold is too high for this file. Lower it.
- MP3 export fails: the encoder could not be downloaded. Choose WAV or check your connection.
Online tool or desktop editor?
Editors such as Audacity include a Truncate Silence effect, and full programs give you waveform editing at the level of a single breath. They are the best choice for finished productions. A browser tool is quicker for everyday jobs, and it keeps the file on your device. It works well as a first pass, after which you can fine-tune in an editor if you need to.
Privacy and how it works
Your browser decodes the file to samples. The page measures the loudness of each ten millisecond slice, finds the stretches that stay below your threshold, and joins the remaining parts with a short fade. Nothing is uploaded, stored or logged. The MP3 encoder loads from a public CDN the first time you choose MP3, but your audio is not part of that request.
Frequently asked questions
How do I remove silence from an audio file for free?
Load your file above. Red areas on the waveform show the silences that will be cut. Adjust the threshold and minimum length if needed, press Play result to listen, then click Remove silence and download. It all runs in your browser.
What threshold should I use?
Start with minus 40 dB, which suits clean speech. In a quiet studio, minus 50 to minus 60 dB catches only true silence. In a noisy room, try minus 30 dB. Click Auto threshold to let the tool choose a value from the quietest part of the file.
What does the shortest silence setting do?
Only gaps at least this long are treated as silence. 500 ms removes noticeable pauses between sentences and leaves normal gaps between words. 200 to 300 ms makes tight, fast-paced edits.
Why keep some air around the speech?
The padding keeps a short moment of quiet at each edge of a cut, so words are not clipped and the rhythm still feels natural. 80 to 150 ms is a good range for speech.
What is the difference between removing and shortening silences?
Removing cuts the silence down to the padding on each side. Shortening cuts every long silence to a fixed length that you choose, for example 300 ms, so the pauses stay but get shorter. Shortening usually sounds more natural for conversation.
What does Trim start and end do?
It removes only the silence at the beginning and the end of the file, and leaves all the pauses inside the recording alone. It is a safe way to clean up a recording before sharing it.
Will it cut my breaths and quiet words?
It can, if they fall under the threshold for longer than the minimum length. Lower the threshold, use a longer minimum, or choose Shorten so that the quiet part is not removed completely. Always preview.
Does it work on music?
It can trim silence at the start and end of a song or remove long gaps, but the music itself usually has no true silence. It is designed for speech such as podcasts, lectures, interviews and voice memos.
Why does it find no silence in my file?
The background noise is probably louder than the threshold. Click Auto threshold, or raise the threshold toward minus 30 dB. A file with constant music or hum under the speech has no real silence to find.
Will cutting cause clicks?
No. A five millisecond fade is applied where the audio is joined, which hides the edit in normal speech.
Is my audio uploaded to a server?
No. The file is decoded and edited on your device. The page downloads an open-source MP3 encoder from a public CDN the first time you choose MP3, but your audio is not sent anywhere.
Can I remove silence from a video?
You can load a video file and process its audio track, but the download is an audio file. Removing audio gaps would also make the video out of sync, so the tool does not return video.
How much shorter will my file get?
The summary shows the new length and the percentage saved. A recorded lecture often shrinks by 10 to 30 percent, and a voice memo with long pauses can shrink even more.
Can I see exactly what was cut?
Yes. The waveform shows the cuts in red, and a table lists each silence with its start, end and length, and how much of it was removed.
Is there a length limit?
No fixed limit, but the audio is decoded in memory, so very long files need more memory. One hour of stereo audio needs roughly 1.3 GB.
Which formats can I save?
MP3 for small files and WAV for uncompressed audio that is best for further editing.
What can I print or save from the page?
Print / save summary prints the settings, the length saved and the waveform with the cuts. Copy summary copies the settings as text. The edited audio is the main download.
Sources and further reading
- AES17, standard method for digital audio engineering: measurement of digital audio equipment, which defines the dBFS scale used for the threshold.
- Audacity Manual, Truncate Silence, a description of the same technique in a desktop editor.
- MDN Web Docs, Web Audio API and AudioBuffer, for how browsers decode and handle audio samples.
- LAME project and the lamejs encoder, the open-source MP3 encoder used for MP3 output.
Sources were consulted in October 2026. Standards and rules can change, so always check the official page for your situation.