How to Convert MP4 to MP3: Bitrate, Trimming and Every Device

An MP4 file is a container, defined by the MPEG-4 Part 14 standard (ISO/IEC 14496-14), that wraps separate streams in one package. A typical phone or camera video holds a video stream compressed with H.264 or H.265, an audio stream compressed with AAC, and a little metadata such as the recording date and rotation. When people say they want to convert MP4 to MP3, they do not want the picture converted at all. They want the audio stream on its own, in a format that every music player, car stereo and podcast app accepts without asking questions.

Here is the catch. The audio inside the MP4 is AAC, not MP3, and the two codecs cannot simply be relabeled. To produce a real MP3, a converter has to decode the AAC back to raw samples and then compress those samples again with an MP3 encoder. Both steps are routine, take a few seconds for a typical clip, and are what every online and desktop tool does under the hood. The video stream is never decoded, which is why audio extraction is much faster than converting a video to another video format of the same length.

Because AAC and MP3 are both lossy, this is a second round of lossy compression on top of the first. That sounds alarming, but in practice it only becomes audible when the second encode is starved of bits. MP3, formally MPEG-1 Audio Layer III, was standardized as ISO/IEC 11172-3 in 1993 and is older and slightly less efficient than AAC, so give it a little more bitrate than the source used. If the video's audio was 128 kbps AAC, an MP3 at 160 or 192 kbps will sound the same to almost anyone on ordinary headphones.

There is a second route that skips the re-encode entirely. An M4A file is the same MPEG-4 container as an MP4, just without the video stream, so a tool can copy the AAC track across bit for bit. Nothing is decoded, nothing is compressed again, and the result sounds exactly like the video did. If the audio is going to an iPhone, a Mac, Apple Music, GarageBand or any modern Android phone, extract MP4 to M4A and you lose nothing at all. Some browser and desktop tools still re-encode M4A output for consistency, but the quality cost at a sensible bitrate is the same tiny one as for MP3.

MP3 earns its place through compatibility. Its last patents expired in 2017, when Fraunhofer IIS ended its licensing program, and decades of hardware were built around it: older car stereos that read USB sticks, cheap MP3 players, some DJ software, voice-over portals and plenty of web forms that accept .mp3 and nothing else. If the file will leave your own devices, or you do not know what will play it, MP3 is the format that will not come back with a support question. For the uncompressed and lossless options, our guide to MP3, WAV, FLAC and AAC compares all four side by side.

Bitrate is the number of kilobits spent on each second of sound, and it alone decides the size of the MP3. Resolution, frame rate and the size of the original video have no effect, because the picture is gone. Multiply bitrate by duration and divide by 8 to get the size: one minute at 128 kbps is about 0.96 MB, at 192 kbps about 1.44 MB and at 320 kbps about 2.4 MB. A one-hour lecture at 128 kbps therefore lands near 58 MB, while a four-minute song at 320 kbps stays under 10 MB.

What you hear at each step is fairly predictable. At 128 kbps a voice sounds natural and music sounds fine on phone speakers and earbuds, but cymbals, applause and reverb tails can turn slightly swishy on good headphones. At 192 kbps most people cannot tell the MP3 from the video's own audio in a blind comparison, which is why it is the usual default on the MP4 to MP3 converter and in many desktop apps. At 320 kbps, the highest bitrate the MP3 standard allows, the file is about two thirds larger again for a difference few listeners notice.

The rule that saves the most disappointment is that a higher bitrate cannot restore what the video never had. Phone recordings typically carry AAC at 96 to 256 kbps, and screen recordings or messaging-app videos are often lower still. Encoding a 96 kbps source at 320 kbps gives you a file three times larger that sounds exactly like the 96 kbps original, artifacts included. When in doubt, check the source bitrate in a media info panel and choose the next step up from it, not the top of the list.

Most people do not want the whole video as audio. They want one song from a two-hour concert recording, a five-minute answer from a webinar, or a sound effect from a game capture. Cutting the clip during extraction is faster and cleaner than extracting everything and editing later, because you never handle the parts you do not need. On the video to audio converter, open Advanced settings and fill in Start at and End at with seconds such as 95 or with minutes and seconds such as 1:35. Leave one field empty to keep the beginning or the end, and play the video first to note the exact times.

Two habits make clips sound finished rather than chopped. Start a second or so before the first word or note, because cuts that land exactly on a sound tend to clip the attack and make it sound abrupt. And if the recording is quiet, as phone videos of a lecture from the back row usually are, tick Normalize volume, which brings the loudness to −16 LUFS, a common target for spoken audio, so the clip plays at a similar level to your other files. If the video carries a second language or a commentary, Audio track picks which one becomes the MP3.

An MP3 made from a video usually arrives with a filename and little else. Music players read the title, artist, album and artwork from ID3 tags stored inside the file, and ID3v2 can embed a cover image as well. On Windows, right-click the MP3, open Properties and then Details to type a title, artist and album. On a Mac, add it to the Music app, choose Get Info and paste artwork into the Artwork tab. Free tag editors such as Mp3tag do the same in bulk, which is worth it when you extract a whole series of talks or episodes at once.

Every platform has some way to get audio out of a video, though few of them produce an MP3 directly. On a Mac, QuickTime Player can save the sound alone with File → Export As → Audio Only, but the result is an M4A, not an MP3. The Music app can then turn it into MP3 once you set Settings → Files → Import Settings to MP3 Encoder and choose File → Convert → Create MP3 Version. On Windows there is no built-in exporter for this, so the common free route is VLC, whose Media → Convert / Save dialog includes an Audio - MP3 profile. Audacity also opens MP4 files once its FFmpeg library is installed, and exports MP3 from there.

Phones are more limited. On an iPhone, the Shortcuts app's Encode Media action has an Audio Only switch, which saves an M4A from a video in your library; it works, but building the shortcut takes a few minutes and the output is still not MP3. Android has no system-wide equivalent, and many of the audio extractor apps in the Play Store are ad-heavy or ask for broad storage permissions. If a clip needs to become text rather than audio, our walkthrough on transcribing audio to text picks up where this one ends.

The browser route is the same on every device: upload the video, pick the bitrate, press Extract to MP3 and download the file. UltKit accepts videos up to 95 MB each, needs no app or account, and deletes both the upload and the MP3 from its servers 30 minutes after processing. Long recordings above 95 MB can usually be shrunk first by lowering the resolution, since the picture quality stops mattering the moment you only want the sound. The steps below summarize the fastest route on each platform, from desktop apps to a phone browser.

Most failed extractions come down to one of four causes, and the error message rarely names it. The most common is a video that simply has no audio stream: some screen recorders capture the picture only unless you switch on the microphone or system sound, and some social apps strip the sound from clips saved without it. The second is copy protection. Films and episodes bought or rented from stores such as the iTunes Store, and downloads saved inside streaming apps, are encrypted with DRM, and no ordinary converter can read the audio inside them. Only convert videos you made yourself or have the right to use, since copyright on the sound track applies whatever the file format.

The other two causes are about timing rather than access. Screen recordings and some phone captures use a variable frame rate, and a few of them also contain gaps or jumps in their timestamps, so the extracted audio can end up a second or two shorter than the video or drift when you line it back up in an editor. Re-saving the video at a constant Frame rate first, with the video converter, usually fixes both. Finally, a file that is a few hundred kilobytes and plays as silence often means the wrong Audio track was picked, which is easy to check by extracting ten seconds of each track.