August 30, 2026

What is an M4A file and how do you transcribe one

An M4A file is the format almost everyone ends up with and almost nobody chose. It arrives from a phone recording, from a voice memo, from an interview captured on a laptop, and it works perfectly until the moment you try to hand it to something else. Then a upload form rejects it, or a transcription service asks for MP3, or a colleague reports that it will not open, and a file that played fine thirty seconds ago becomes an obstacle.

The confusion is worth clearing up properly, because most advice about M4A is a conversion instruction dressed up as an explanation.

What is an M4A file?

An M4A file is an audio file stored in the MPEG-4 container, holding audio only rather than audio and video. The audio inside is usually encoded with AAC, a lossy codec, though the same container can also hold Apple Lossless. The extension is essentially a courtesy: it tells software and humans that this particular MPEG-4 file has no video track in it, so an audio player should handle it rather than a video player.

That covers the definition. Everything confusing about M4A follows from the fact that it describes a container rather than a compression method.

Container and codec are not the same thing

Thirty seconds on this distinction explains most M4A problems.

A container is a box. It specifies how the data is arranged in the file, where the metadata sits, how playback software should find the audio track. A codec is what was done to the audio before it went in the box. MP3 is unusual in that the file format and the codec are effectively the same thing, which is why nobody has to think about this when dealing with MP3, and why the habit of not thinking about it carries over badly.

With M4A the box tells you very little about the contents. Two M4A files can contain audio compressed in entirely different ways and at wildly different quality levels. One might be a lossless copy of a studio recording. The other might be a phone memo compressed hard enough to save space on a full device. Same extension, same box, different contents.

Which means the question is my file M4A is rarely the useful question. The useful question is what is inside it and what does the receiving system actually accept.

Why your recordings are M4A in the first place

Because the default made sense at the time and nobody revisited it.

AAC in an MPEG-4 container gives better quality than MP3 at the same bitrate, handles speech well at low bitrates, is broadly supported across modern devices, and has none of the licensing history that made MP3 awkward for a period. For a phone recording app that has to store hours of audio on a device with finite storage, it is a straightforwardly good choice.

Apple documents the workflow plainly enough. Recordings can be exported from Voice Memos to the Files app, and the Mac guide frames the app as somewhere to record meetings, lectures and song ideas. Notice what both descriptions have in common. They describe capture. Neither describes what happens to the recording after it exists, and that gap is where the format question starts to bite.

M4A against the formats it gets compared to

Most comparisons of audio formats are written for music. For speech recordings the trade-offs land differently, and the differences that matter shrink to a shortlist.

Format	Container and codec	Typical use	Speech at low bitrate	Where it causes friction
M4A	MPEG-4 container, usually AAC	Phone recordings, voice memos, exported interviews	Strong	Older upload forms, some legacy tools
MP3	MP3 both	Universal fallback	Adequate	Almost none, hence its use as a fallback
WAV	Uncompressed PCM	Editing, archival masters	No compression, so no artefacts	Enormous files, impractical to store or send
AIFF	Uncompressed, Apple origin	Editing on Mac	Same as WAV	Same as WAV, plus narrower support

Read across the table and the picture is unremarkable. M4A is the technically better choice for speech at small file sizes, MP3 is the more universally accepted one, and the uncompressed formats are only relevant if the recording is going to be edited seriously rather than listened to or read.

None of which resolves the situation you are actually in, which is that the file already exists, in M4A, and something downstream is refusing it.

Where M4A actually breaks

Not in playback. Playback is fine almost everywhere in 2026.

It breaks at upload boundaries. Older content management systems, some learning platforms, certain podcast hosts and a long tail of internal corporate tools accept MP3 and WAV and nothing else, usually because the accepted list was written a decade ago and nobody has revisited it either.

It breaks in editing software that will import the file but insists on conforming it first, which is fine once and irritating across forty files.

And it breaks in transcription workflows, which is the case most people actually hit, because the recording that most needs to become text is precisely the phone memo that is sitting in M4A.

Converting M4A, and why it is usually the wrong first move

The reflex is to convert to MP3. Sometimes that is correct. Frequently it makes things worse in a way that is invisible until later.

Converting AAC to MP3 is a lossy to lossy conversion. The audio is decompressed and then recompressed using a different method, and each stage discards information according to its own assumptions. For music this is audible under scrutiny. For speech at typical voice memo bitrates it is often not audible at all, but it is not free either, and it is particularly unhelpful if the recording was already made in poor conditions, because the artefacts stack on top of the room noise rather than replacing it.

There is also a practical cost that nobody counts. Converting one file takes a minute. Converting two hundred takes an afternoon, produces a duplicate library at twice the storage, and leaves you with two versions of everything and no memory of which is which.

The rule of thumb worth adopting: convert when a specific system requires a specific format, and not otherwise. Convert at the boundary, not in advance.

Transcribing M4A directly

For transcription specifically, conversion is almost always unnecessary, because the format was never the real constraint. Speech recognition operates on decoded audio, and decoding AAC is trivial for anything built in the last decade.

So the sensible path is to hand over the M4A as it is. A tool that offers M4A audio transcription without a conversion step removes the entire question, and Vomo also places no cap on the length of an individual file, which matters more than it sounds. Long recordings are exactly the ones people avoid dealing with, and a two hour interview that has to be manually sliced into ten minute pieces before it can be processed will simply not get processed.

Transcripts can then be taken out as text, DOCX, PDF, SRT or Markdown depending on where they are going next.

The backlog underneath the format question

The situation that actually exists on most phones has nothing to do with formats.

There are two hundred recordings. Some are from three years ago. They are named by date and time, which conveys nothing, and the only way to find out what any of them contains is to play it. So none of them get played, the collection grows, and its usefulness drifts toward zero while the sense of obligation attached to it grows.

Format has nothing to do with this. Converting all two hundred to MP3 would leave you with two hundred equally unsearchable files in a different box.

What changes the situation is text, because text can be searched, and because a transcript of a four minute memo can be scanned in ten seconds where the memo itself takes four minutes. Two hundred recordings is perhaps eight hours of audio and perhaps forty thousand words, which is a readable quantity.

One honest note on scale. A backlog of that size will not clear on a free weekly allowance. Vomo’s free tier provides thirty minutes of transcription per week, which suits ongoing use and does not suit a three year archive, so a bulk clearout is a paid tier exercise. Better to know that now than at file eleven.

A short checklist

For anyone dealing with M4A files regularly, five rules cover almost every case.

Do not convert by default. Establish first whether the destination actually rejects M4A, because most no longer do.

If conversion is required, convert a copy and keep the original. The original is the only lossless thing you have.

For transcription, skip conversion entirely and submit the M4A.

Rename files at the point of capture, not later. A single spoken sentence at the start of a recording is worth more than any filing system applied afterwards.

And set aside one session for the backlog rather than adding it to a list of things to get to. Most people have added it to that list more than once already.