Runs on your device · Your file never leaves your device.
Pull the audio out of a video file and save it as an MP3 — useful for lectures, interviews, podcasts and voice recordings you were sent as video.
The video is decoded in your browser and the audio re-encoded to MP3, so the file never leaves your device. That matters for recordings of meetings and consultations.
Longer videos take longer to process, and very large files are limited by your device’s available memory rather than by any server quota.
Not for: Pulling several separate sound tracks out of one film — one MP3 carries one sound track, the film’s main one.
The Quality slider runs from 48 to 256 kbps and starts at 128. The MP3 is written at a constant bitrate, so its size depends only on the setting and the length: 64 kbps is about 0.5 MB per minute, 128 kbps about 1 MB, 192 kbps about 1.4 MB. An hour at 64 kbps is under 30 MB.
The sound inside an MP4 is almost always compressed already, usually as AAC, so this is a second compression. A higher setting keeps more of what is there, but cannot bring back what the video’s own encoding removed. For talk, 64 or 96 kbps is enough; for music, 160 kbps or more.
In an MP3 the bitrate alone sets the size, so “Downmix to mono” at the same setting gives a file of the same size. What changes is how the bits are used: the channels are averaged into one, and the whole bitrate goes to that channel. A voice at 48 or 64 kbps sounds cleaner in mono than at the same setting in stereo, and that low setting is where the saving comes from.
Mono also helps when the voice was recorded on one side only, as with some clip-on microphones: after averaging it plays in both ears, somewhat quieter. Keep stereo for music.
The browser’s own decoder unpacks the audio track into raw samples; the picture is not decoded. An MP3 encoder running in the page then writes the file. If the browser cannot decode the track, for example AC-3 or DTS sound, the audio in MKV, AVI or VOB files, or AAC in a browser built without it, the site’s video engine extracts it first as 44.1 kHz stereo and the same encoder takes over. The MP3 takes the video’s file name and carries no title or artist tags. For a cut to a tenth of a second, with a short fade, use the Cut tab instead of the whole-second trim slider.
“This file has no audio this browser can read” appears only after both decoders have tried. The usual causes are a video recorded without sound, as many screen recordings are, or a damaged or partly downloaded file.
Memory is the real limit. The whole video and its unpacked sound are held at once, and one hour of stereo unpacks to more than a gigabyte of raw samples before any MP3 is written. If a long video stops your phone, cut it into pieces with /trim-video, then add the pieces to /merge-audio, which reads them one at a time and writes a single MP3.
This page in other languages: العربية · Deutsch · Français · Español · Русский · Português · 中文 · हिन्दी · Türkçe · עברית · فارسی · اردو · Bahasa Indonesia · বাংলা