Runs on your device · Your file never leaves your device.
Pick the start and end on a drawn waveform, listen to just that stretch, and save it as MP3 or WAV. Ringtones, a quote from a lecture, the clean minute of a voice memo, the intro of a song — anything that only needs part of a recording.
The file is decoded in your browser and the cut is encoded right there, so a private recording stays on your device. Drag the handles with a finger or a mouse, or type exact times to a tenth of a second; the quick picks take the first or last 30 seconds, or drop the first or last 10, in one tap. A fade in or out of up to three seconds keeps the cut from starting or ending abruptly.
It takes any audio the browser can decode — MP3, M4A, WAV, OGG, FLAC, OPUS — and video files too, from which it reads the audio track. The output is a new file; the original is untouched.
Not for: Removing a section from the middle of a recording — this tool keeps one continuous range; cut the two halves separately, then join them with Merge audio.
The waveform shows the whole file at once and has no zoom, so on a long recording a small movement of the handle covers several seconds. Use it to get close, then finish in the Start and End fields. They accept 75, 1:15, 1:15.5 or 1:01:15, and a comma works as the decimal mark. A time is applied when you press Enter or leave the field; anything the field cannot read is put back, with a hint to type a time like 1:05.5.
On a computer, move to a handle with the Tab key and use the arrow keys: each press moves it a tenth of a second, or a full second with Shift held, and Home and End send it to the start or end of the file. The selection cannot be shorter than a tenth of a second, and the start cannot pass the end.
For a precise cut deep inside a long recording, cut generously first and save that part as WAV, then open the WAV and cut again. The shorter file fills the whole waveform, and WAV adds no loss of its own in between.
The selection is decoded and encoded again into a new file, named after the original with “-cut” added. Tags are not carried over, so a song’s title, artist and cover picture will not be in the cut. The fades are straight ramps in volume, applied to the cut only, and Play selection fades in the same way, so the preview matches what is saved.
Every MP3 encoder adds a few hundredths of a second of silence at the start and end. An MP3 made here carries a header saying exactly how much, so a player that reads it skips that silence, and two cuts joined back to back on /merge-audio meet without a gap.
Each MP3 step still compresses the sound again. When you cut two halves to put back together, saving both as WAV and choosing MP3 only for the final merged file means the sound is encoded once.
This page in other languages: العربية · Deutsch · Français · Español · Русский · Português · 中文 · हिन्दी · Türkçe · עברית · فارسی · اردو · Bahasa Indonesia · বাংলা