Silence Trimmer
Detect and cut the silent parts of an audio file — start, end or the gaps in between — and download a trimmed 16-bit WAV.
Drag & drop an audio file here, or click to browse
No uploads. Your audio stays on your device.
The shaded blocks are the silences that get cut out; everything else is kept.
How it works
Drop an MP3, WAV, M4A, OGG, FLAC or WebM file and the Web Audio API decodes it inside this tab — nothing is uploaded, so voice memos, interviews and unreleased takes never leave your device. The decoded samples are scanned in 10 ms windows: any window whose RMS level sits below the silence threshold counts as silence, and runs of those windows that last longer than your minimum become candidates for the chop.
Silence threshold is measured in dBFS, where 0 dBFS is the digital maximum. −45 dBFS is a good default for a reasonably clean recording; raise it towards −30 for a noisy room, drop it towards −60 if quiet speech is being eaten. Minimum silence stops natural pauses between words from being cut, and keep padding leaves a little breathing room at each edge of every cut so the result never sounds clipped. Choose whether to trim only the dead air at the start and end, or to close the gaps in the middle too. The result plays back instantly and downloads as an uncompressed 16-bit WAV at the original sample rate and channel count.
Frequently asked questions
How does the tool decide what counts as silence?
The decoded audio is scanned in 10 ms windows and the RMS level of each window is compared with the silence threshold you set, measured in dBFS where 0 dBFS is the digital maximum. Any window quieter than the threshold is marked silent, and only runs of silent windows that last longer than your minimum silence length are treated as real pauses — that stops the natural micro-gaps between words being chopped out. Every cut then keeps a slice of padding at each edge so the result never sounds abruptly clipped. Because the waveform shades exactly which regions will go before you download anything, you can nudge the three settings and watch the effect immediately.
What silence threshold should I use?
−45 dBFS is a sensible default for a reasonably clean recording made with a decent microphone. If the tool leaves long pauses in place, the room tone is louder than the threshold — raise it towards −35 or −30 dBFS so that background hiss counts as silence. If quiet speech, breaths or the tails of instruments are being eaten, go the other way, down towards −55 or −60 dBFS. Pair the threshold with the minimum silence length: 300 ms keeps speech sounding natural, while 800–1000 ms only removes the obviously dead stretches. Raising the keep-padding from 100 ms to 200–300 ms is the quickest fix if the edits sound too tight.
Is my audio uploaded anywhere, and what do I get back?
Nothing is uploaded. The file is decoded with the Web Audio API inside this tab, the silence analysis runs on the decoded samples in memory and the trimmed file is assembled in the page — no server, no account, no shareable link created by accident. That matters for interviews, voice memos, therapy notes and unreleased music. The download is an uncompressed 16-bit WAV at the original sample rate and channel count, so there is no second round of lossy compression; you can drop it straight into an editor or convert it afterwards if you need MP3.