Video to MP3

Take the audio track out of a video and save it as an MP3 or a WAV, without the file leaving your machine.

Runs on your deviceVideoWorkingNo account needed

What this tool does

Video to MP3 hands the file to the browser's own audio decoder, reads the samples that come back, and writes them out again as either an MP3 or a 16-bit PCM WAV. The decode, the trim, the mono fold and the LAME encode all run in this tab, so the video is not uploaded anywhere — the MP3 encoder is a script served alongside the page rather than a service on the other end of a connection.

Which files work is decided by the browser rather than by this page. An MP4 carrying AAC decodes straight away in Chrome, Edge and Safari, and in Firefox wherever the system supplies an AAC decoder. When the container defeats the decoder but the browser can still play it — a Matroska file is the usual case — the tool falls back to playing the video through an audio graph that is never connected to the speakers and recording the result as it goes. That works, but it runs in real time (a forty-minute lecture takes forty minutes) and the capture itself passes through Opus on the way. An AVI, WMV, FLV or MPEG-PS file gets neither route, because no browser will open it at all: the tool says which container it is and sends you to Convert Video rather than stalling. For MP4, MOV, WebM and MKV the header is parsed before any of that, so a video with no audio track is refused immediately instead of after a long silent wait.

Two things it will not do, and both are worth knowing before you start. The audio is decoded and re-encoded rather than copied out of the container, so an MP3 made from an AAC soundtrack is a second lossy generation — pick WAV when you intend to edit further, accepting that an hour of 48 kHz stereo lands near 660 MB. And the MP3 path carries at most two channels, so a surround mix loses everything after left and right, while no ID3 tags are written at all: the result arrives with a filename and no title, artist or album field set.

How to use it

  1. Choose one or more videos. MP4, MOV, M4V, WebM, MKV, AVI, MPEG and the other common containers can all be selected. Each file is handled in turn and saved as its own audio file.
  2. Pick MP3 or WAV. MP3 for something to keep or send on. WAV writes 16-bit PCM at whatever sample rate the decoder reported, which is much larger but adds no further loss.
  3. Set the bitrate and the channels. MP3 bitrates run from 64 kbps for speech through 192 kbps up to 320 kbps. Mono averages the channels together into one.
  4. Trim with Start at and End at. Leave both blank for the whole track, or give plain seconds (90), mm:ss (1:23) or hh:mm:ss (0:01:23) to keep a single section.

Processing & privacy

Runs on your device. Your file is processed directly in this browser.

Common use cases

Keeping the audio of a recorded talk

A conference recording is a large thing to carry around when only the speech matters. 96 or 128 kbps in mono stays perfectly listenable at a fraction of the size.

Lifting one section out of a long clip

Set Start at and End at around the part you want and only that range is encoded, so there is no second pass through an audio editor afterwards.

Getting audio from video for a transcript

Most transcription services want an audio file rather than a video, and extracting first cuts what you hand them down to the part they actually read.

Salvaging sound from a phone recording

Someone filmed a meeting on a handset and the picture is useless, but the sound is the only record of what was agreed.

Questions

Is my video uploaded?
No. The decode and the encode both happen in this browser tab, and the MP3 encoder is loaded as a script with the page. That is also why the practical size limit is your device's memory rather than an upload ceiling.
Can I do mp4 to mp3 without losing quality?
Not quite. The track is decoded and re-encoded rather than copied out of the container, so an MP3 made from an MP4's AAC audio is a second lossy pass. At 192 kbps and above the difference is hard to hear; WAV avoids the second pass entirely, at the cost of size.
Why is it taking as long as the video is long?
The browser could not decode that container directly, so the file is being played through a silent audio graph and recorded as it goes, which is real time by nature. Converting the video to MP4 or WebM first avoids that path.
Does choosing mono make the file smaller?
The option is labelled “half the size”, and for WAV that is right — half the samples, half the bytes. For MP3 it is not: the bitrate you pick fixes the size, so mono spends the whole of that bitrate on one channel and the result sounds cleaner rather than shrinking.
It says no sound could be read. What went wrong?
One of three things: the video has no audio track, the track uses a codec this browser cannot decode, or everything in it sits below −80 dBFS. In each case nothing is written, because a silent MP3 would look like a successful result.
What happens to a 5.1 soundtrack?
The MP3 path keeps the first two channels and drops the rest, since MP3 here is written as mono or stereo. WAV keeps every channel the decoder returned, so choose that when the surround mix matters.
Why did the sample rate change?
The MP3 encoder only accepts a fixed set of rates, from 8 kHz up to 48 kHz. Audio arriving at anything else — 96 kHz, for instance — is resampled to the nearest one it supports before encoding. WAV is written at the original rate.

Related tools