Local processing
Free

MP4 to MP3 Converter Extracted in this tab, not uploaded

This takes the audio track out of an MP4 file you already have and writes it to an MP3 at 128 kbps, 44.1 kHz. It is extraction, not restoration, so the result sounds like the audio in the video and no better than that. It is also not a downloader: it never fetches anything from YouTube or any other website. Point it at a file on your device and the whole job happens inside this browser tab.

  • 128 kbps MP3
  • Nothing uploaded
  • 20 files per batch

Convert MP4 to MP3

MP4, MOV, M4V, MKV, AVI, WMV, WebM, TS, MTS and more

Up to 20 clips at once

Privacy-first No uploads Processed locally

Privacy-first. No uploads. If this helps, share it.

  • The recording stays with you

    A WebAssembly build of FFmpeg reads the file inside this tab and writes the MP3 there too. Speech carries names, addresses and whole conversations, so an interview or a recorded call is exactly the sort of thing worth not handing to a stranger's server.

  • A plain 128 kbps MP3

    Every export is 128 kbps at 44.1 kHz, which is the sensible middle for talking: clear enough for a lecture or a podcast, small enough to mail. No watermark, no spoken intro, no tag pointing back at us.

  • 20 files at a time, 1 GB each

    Queue a folder of lecture recordings and walk away. The browser build runs on one thread, so files are handled in order rather than side by side, and each MP3 can be saved the moment its own turn ends. The finished set also downloads as a single zip.

Quick answer

How to convert an MP4 to MP3

Add the MP4, make sure the output is set to MP3, and the audio stream is decoded and re-encoded to a 128 kbps MP3 here on your machine. The picture is not converted, it is skipped, so this is much lighter work than re-encoding a video. Your original MP4 is untouched, and the file is never sent anywhere.

  1. Add your MP4
  2. Keep the output on MP3
  3. Save the audio
The formats

MP4 next to MP3, side by side

An MP4 is a box with several things inside it. An MP3 is one stream and a little bit of tag data. Going from one to the other means keeping a single track and throwing the rest away, so it helps to see exactly which parts survive.

What changesMP4 sourceMP3 result
ContainerMP4 holds video, audio, chapters, subtitles and metadata togetherMP3 holds one audio stream plus ID3 tags, nothing else
Video streamUsually H.264, sometimes HEVCGone. Thumbnails, subtitles and chapter marks go with it
Audio codecUsually AAC, occasionally AC-3 or already MP3Re-encoded to MP3 at 128 kbps, 44.1 kHz, every time
Where it playsVideo players, phones, editors, upload formsMusic apps, podcast players, car stereos, transcription tools, audio editors
File sizeMostly the picture track, which is where nearly all the bytes sitAround 1 MB per minute, since the bitrate is fixed whatever the source was

The To dropdown above also offers MP4, so if you decide you wanted the video after all you can switch there without leaving the page. The page is not fussy about what goes in either: drop a MOV, an MKV or a WebM instead of an MP4 and the audio comes out the same way, with the result named for the file that actually arrived.

The tradeoff

What you gain, and what you give up

What you gain is a file that behaves like audio. It queues in a podcast app, plays through a car stereo over Bluetooth, feeds a transcription service without dragging a picture track through the upload, and takes a fraction of the space. What you give up is a generation of quality. The audio inside an MP4 has already been compressed once, usually as AAC, and writing it out as MP3 compresses it a second time from that already lossy copy. For speech the difference is hard to notice. On music, on anything with cymbals or heavy reverb, a careful listener on decent headphones can hear it.

Nothing here can be better than what is in the file. If the video was recorded through a laptop microphone in a room with an air conditioner, the MP3 has the laptop microphone and the air conditioner in it. Extraction copies the sound across. It does not clean it up.

Where it lands

What an MP3 opens in, and what it will not do

MP3 is the format with the fewest arguments left in it. Phones, desktop operating systems, car head units, podcast hosts and nearly every transcription or note taking service will take one without complaint. AAC inside the MP4 is the better codec at the same bitrate, but MP3 is the one that opens everywhere without a conversation, and that is usually what you want from a file you are about to hand to somebody else.

  • Nothing has to be installed to play it. Phones and laptops, Bluetooth speakers and car stereos, podcast apps and RSS feeds, transcription and note taking tools, audio editors and any old MP3 player still in a drawer.

  • It will not go anywhere that expects a picture. An MP3 handed to a video upload slot is rejected, so keep the MP4 as well whenever the platform wants video.

  • It cannot be cut here. This tool converts whole files. Taking the middle four minutes out of an hour, fading an intro or stripping background noise needs an audio editor after the export.

Three steps, all of them local

The file goes nowhere.

  1. 01

    Add the videos

    Drop MP4 files in or browse for them. A batch holds 20 files and each one can be 1 GB, which is enough for lecture captures, Zoom exports and phone recordings.

  2. 02

    Check the output is MP3

    The dropdown decides what comes out. On MP3 the audio stream is decoded and re-encoded at 128 kbps, 44.1 kHz, and the picture is ignored entirely.

  3. 03

    Take the audio

    Save each MP3 as it appears or grab the batch as a zip. Your MP4 files are still sitting where you left them, unchanged.

Why pull the audio out at all

Plenty of video gets watched once and listened to five times. When that is the pattern, carrying the picture around is a cost with no return.

Some video was only ever audio

A two person podcast filmed for the clips, a lecture where the camera points at a whiteboard for ninety minutes, an interview over Zoom, a voice memo your phone happened to film. The information is in the talking. The picture is packaging.

Audio apps want an audio file

Podcast hosts, transcription services, dictation tools and playlist apps are built around audio formats. Hand them an MP4 and you get a rejection or a slow upload of data the service is going to discard anyway.

It plays where video will not

Car stereos, cheap Bluetooth speakers, running watches and offline music players all handle MP3 and mostly do not handle video files. Extracting is how a recorded talk gets into a commute.

Listening back is a different job

Music practice is a good example: loop a rehearsal recording, learn a part by ear, compare two takes. All of that is easier once the sound is a standalone file rather than something you scrub through in a video player.

Privacy

A voice recording is not a neutral file

Think about what tends to be in one. A client call with figures in it. A therapy session. A lecture the university has not published. An interview where a source is identifiable from thirty seconds of speech. Uploading that to a converter means a stranger holds a copy, and audio is easy to search once someone runs it through speech recognition.

Here there is no upload leg at all. The FFmpeg build runs as WebAssembly inside the page, the MP3 is written in your browser's memory and handed straight to your downloads folder. We never see a byte of it, nothing waits in a processing queue, no temporary copy is made on our side, and since there is nothing to sign in to, there is nowhere for a record of your files to exist.

What that means in practice

  • The audio and the file name both stay on your device
  • No account, no watermark, no audio tag added to the MP3
  • The work runs on your own processor, so a three hour recording is not instant
  • A 1 GB file has to fit in browser memory, which favours a laptop over a phone

When the MP3 is not what you expected

Frequently asked questions

Add the MP4, make sure the output is set to MP3, and save the file that comes back. The audio is decoded and re-encoded at 128 kbps inside your browser, with no upload and no account.

Reviewed in August 2026. Output is 128 kbps, 44.1 kHz MP3.