Runs on your device · Your file never leaves your device.
Add the recordings, put them in order, and get back one file. Voice memos recorded in pieces, a lecture split across several files, chapters of an audiobook, or a playlist you want as a single MP3 for a car stereo or a slideshow.
The files need not match: a stereo MP3 at 44.1 kHz, a mono WAV at 48 kHz and an M4A from a phone are all brought to one sample rate and channel count before they are joined. Leave up to five seconds of silence between tracks, or switch on a one-second crossfade so the end of each track melts into the next.
Everything runs in your browser. Each file is decoded in turn and handed to the encoder straight away, so the memory in use is one track at a time rather than all of them — a long merge is not limited by holding every file at once.
Not for: Mixing tracks on top of each other — files are joined one after another, not layered.
The result is stereo whenever any file in the list is stereo, or does not say in its header; it is mono only when every file is known to be mono, or when you switch on Downmix to mono. A file with more than two channels, such as a 5.1 soundtrack, is mixed down to stereo with its centre included. The merge runs at the first file’s sample rate, and the others are converted to it.
The merged file has no tags and no chapter marks, and is saved as merged.mp3 or merged.wav. For an audiobook or a series of lectures, a pause of one or two seconds keeps the boundaries easy to hear.
The pause is silence of 0 to 5 seconds, in half-second steps, placed between tracks but not before the first or after the last. The crossfade overlaps the last second of each track with the first second of the next, lowering one along a curve while the other rises, so the overall level does not dip in the middle. Each crossfaded join makes the result one second shorter. While the crossfade is on, the pause setting is hidden, because the two cannot be combined.
A crossfade suits music. In speech it can blur the last word of one file into the first word of the next, so a short pause is usually clearer.
Volumes are not evened out. A quiet phone recording stays quiet next to a loud music file, so check the levels when you listen to the result.
Most files are decoded only once, during the merge. If one cannot be read, the merge stops, names the file and marks it in the list; neither the browser nor the site’s video engine found sound in it, usually because it is damaged, partly downloaded, or a video recorded without sound.
MP3 parts cut on /mp3-cutter carry a header that tells the decoder how much encoder silence to skip, so they join without a gap in a browser that reads that header, as Chrome does.
This page in other languages: العربية · Deutsch · Français · Español · Русский · Português · 中文 · हिन्दी · Türkçe · עברית · فارسی · اردو · Bahasa Indonesia · বাংলা