Automatically remove long silent passages from recordings (dictation, podcasts, captures) - via level analysis, locally in the browser, no upload.
Running locally on your device ...
Running locally on your device ...
0%
Your files never left your device
No. Everything runs in your browser - your file never leaves your device. How this is verifiable
Dictation, interviews and lecture recordings often consist substantially of pauses. This tool measures the level in short analysis windows and cuts out silent passages longer than the chosen minimum - shorter speech pauses stay untouched so the natural rhythm survives.
At every cut, 0.2 seconds of the silence is kept. The result never sounds chopped: word starts and ends keep their air, only the long emptiness in between disappears. Sensitivity decides what counts as silent: -30 dB removes quiet rustling too, -50 dB only near-absolute silence.
The analysis runs fully locally over the decoded stream - even for long recordings the memory footprint stays small, because discarded silent passages are never buffered. Nothing is uploaded.
| Input formats | MP3, WAV, M4A, AAC, OGG, OPUS, FLAC |
|---|---|
| Auto-conversion | MP4 |
| Output formats | WAV, M4A |
| Batch processing | No |
| Processing | Locally in your browser (WebCodecs) |
| File upload | None |
Limitations: Honest per-device length cap (desktop about 20 minutes, tablet 15, phone 10; more with plenty of memory) - longer files are clearly refused, never silently truncated. Detection is level-based (RMS in roughly 20-100 ms windows): constant loud noise (fans, traffic) does not count as silence - switch sensitivity to -30 dB in that case. 0.2 s of silence is kept as padding at every cut.
| Level | What gets removed |
|---|---|
| Strict (-30 dB) | removes quiet parts too |
| Standard (-40 dB) | suits normal recordings |
| Careful (-50 dB) | true silence only |
No, analysis and cutting run entirely locally in the browser.
No: short speech pauses are fully kept, and 0.2 seconds of silence remains as natural padding at every removed passage.
Anything under the chosen level threshold: -40 dB suits normal recordings, -30 dB removes very quiet parts too, -50 dB only near-absolute silence.
With constant, clearly audible noise the level often sits above the threshold - choose the strict setting (-30 dB) then. It is not a noise reducer.
Depends on the material: dictation with long thinking pauses often loses a third or more of its length, dense music virtually nothing.
Trim audio · Merge audio files · Stereo to mono · Normalize audio