By ShinobiTools Team · Last updated: August 2026
Podcasters and voice-note recorders both ask for this: cut the long silences and dead air out of a recording without doing it by hand, one gap at a time. Working with a plain voice recording instead of a song? Remove silence from the audio with this tool, built for spoken files.
Drop a file in, pick a sensitivity, and get an MP3 back with the dead air gone — a short pause is always left so sentences don't run into each other.
Full-quality stems. No account, no per-track fee. Same model and same 192 kbps (or lossless WAV) as the paid tier — tracks up to 15 minutes.
It runs the same detection in a phone browser as on a desktop, with no separate mobile version and no difference between the free and paid result.
Files up to 120 MB or an hour of audio are free, with no account.
Two things have to both be true before a gap gets cut: it has to be quieter than the sensitivity threshold, and it has to last longer than about 0.7 seconds. That 0.7s floor is deliberate — ordinary pauses for breath between sentences are shorter than that, and cutting those out is exactly what makes a trimmed recording sound rushed and unnatural. Only genuine dead air passes both bars.
This only removes dead air — it keeps the pace of what you actually said. If you also need a specific piece trimmed off the start or end, that's the audio cutter. Recorded several takes and need them in one file? Use the audio joiner first, then run the result through here. Want the vocal separated from music underneath? That's the splitter on the home page. It handles video files too: drop in an MP4 and its soundtrack is trimmed for dead air, and you get the gap-free audio back as an MP3 to lay under your picture.
Yes — no account, no e-mail, no watermark and no daily cap. Files can be up to 120 MB or an hour of audio on the free tier.
Anything quieter than the sensitivity threshold, for longer than about 0.7 seconds. That 0.7s floor is deliberate: ordinary pauses for breath between sentences are shorter than that, so they are left alone. Only the long dead-air gaps get cut, not constant background hiss.
No. A short pause is always left behind at every cut — around three quarters of a second — so sentences never run into each other. We tested this against a real recording sample-by-sample: every word came out byte-for-byte identical to the original, just with the dead air around it shortened.
Yes — the Sensitivity slider. Gentle only removes very obvious dead air; Aggressive also catches quieter room tone, which helps on a noisier room recording. Normal is the default and works for most recordings.
No. It measures loudness over time and cuts the stretches that stay below the threshold for long enough. That's predictable, and it means the words themselves are never touched.
Then almost nothing changes — you get the same recording back, just re-encoded to MP3, with only the leading and trailing silence trimmed.
MP3, WAV, M4A, FLAC, OGG and video files all go through the same sensitivity slider. A video's audio track is pulled out first and the silences are cut from that, so the download is an MP3, not the video itself.
Not yet — download the trimmed MP3 to hear the result; you can run it again with a different sensitivity as often as you like.
CapCut and Premiere Pro have silence removal inside a full video editor, so you install an app and cut on a timeline. This does just the one job in the browser: upload, pick a sensitivity, download the MP3. Use it for audio; for cutting the video itself, stay in your editor.
No. Your upload is deleted the moment the trimmed file has been sent back. Nothing is shared or used to train anything — see privacy.
ShinobiTools · KvK 42147182 · info@shinobitools.com