M4A to Text: Transcribing iPhone Voice Memos (and other M4A files)

First: get the M4A off your iPhone (3 methods that work)

All three methods below preserve the M4A format — none of them silently convert to a lossier format on the way over.

AirDrop to a Mac (fastest)

Open Voice Memos → tap the three dots on a recording → Share → AirDrop → pick your Mac. The file arrives in Downloads as the original M4A. This is the only zero-friction method if you live in the Apple ecosystem.

Save to Files app → iCloud Drive (works on Windows too)

Same Share menu → Save to Files → pick iCloud Drive. From a Mac or Windows PC with iCloud, the file appears in the iCloud Drive folder within a minute or two. Best when you need the file on a non-Apple machine.

Email or Messages to yourself (no extra apps)

Share → Mail or Messages → send to your own address. Slowest for long recordings (email providers cap attachments at ~25 MB), but requires no setup. Useful in a pinch when you don’t have AirDrop or iCloud handy.

Does iPhone already transcribe Voice Memos? (Yes, with limits)

iOS 18 added on-device transcription to the Voice Memos app. Open any recording, tap the transcript icon (looks like a text bubble), and you get a synced transcript next to the waveform. It’s genuinely useful and runs without an internet connection.

What works:

• Short clips, clean audio, one speaker
• English (officially the only supported language at launch)
• Recordings made elsewhere, imported into Voice Memos
• Tap-to-jump from transcript text to audio position

What doesn’t:

• Recordings longer than ~30 minutes (transcript stops generating)
• Multi-speaker meetings — there’s no speaker labelling
• Most non-English audio
• Exporting the transcript as a file (you can copy-paste only)

When to use it: a 3-minute note to yourself in English. When to switch:anything you need to share, edit, or process — including most of the recordings you’d actually want a transcript of.

The 4 ways to transcribe M4A files

1. iPhone Voice Memos built-in (free, short English clips)

Covered above. Best for short personal notes; not useful for anything you need to share or process.

2. DeluxeScribe or similar AI service

Upload the M4A, pick the language (or leave on auto-detect), wait 5–10 minutes per hour of audio. Returns a clickable transcript with speaker labels, exportable as TXT, DOCX, PDF, SRT, VTT, or JSON. The free tier (60 minutes, no card) covers a few hour-long interviews without paying.

3. Self-hosted Whisper

Free, private, accurate. Requires Python and patience on a CPU; fast on a recent GPU. M4A is supported directly:

pip install openai-whisper
whisper voice-memo.m4a --model large-v3 --language English

For speaker labels, pair Whisper with whisper-diarization — it adds a setup step but produces output comparable to commercial services.

4. Apple’s Live Captions accessibility feature

On iPhone (iOS 16+) and Mac (macOS Ventura+), Live Captions transcribes any audio playing on the device in real time — including audio from another app or a video call. Useful for getting captions from a meeting you’re watching live, not for transcribing existing recordings (it doesn’t save the transcript by default).

What is an M4A file? (Quick explainer)

M4A is the audio-only variant of MP4. The container is the same as MP4 video; it just contains an audio track (usually AAC, sometimes Apple Lossless) and no video track. The .m4aextension is a convention to signal “this MP4 has audio only” — internally it’s identical to an MP4 with the video stripped out.

Why iPhone uses it: AAC at 128 kbps sounds roughly equivalent to MP3 at 192 kbps, so you store ~30% smaller files for the same perceived quality. Apple defaulted to AAC across iTunes and the iPhone in 2003 and has stuck with it.

Why renaming .m4a → .mp3 doesn’t work: the extension lies but the file bytes don’t. Any MP3 decoder will reject it. To actually convert, re-encode with ffmpeg:

ffmpeg -i input.m4a -acodec libmp3lame -q:a 4 output.mp3

But for transcription, don’t bother — upload the M4A directly.

Accuracy by recording scenario

Common M4A errors and what they mean

“Unsupported file”

Often the file is actually .m4p (DRM-protected, from old iTunes Store purchases), or the extension is .m4a but the contents are something else. Check with ffprobe yourfile.m4a.

File won’t upload

Either the file is over the service’s size limit, or the MIME type is being misreported by the browser. Try uploading from a desktop browser instead of mobile; if that still fails, re-export from Voice Memos at a lower bitrate in Settings → Voice Memos → Audio Quality.

Transcription returns empty text

The audio track is silent or the codec inside the container isn’t actually audio. Play the file locally first — if you hear nothing, the recording itself failed.

Wrong language detected

Voice Memos sometimes tags the file with the device’s UI language rather than the spoken language. Set the language manually in the transcription settings.

When DeluxeScribe is the right fit

The right answer is uswhen you have a long recording, multiple speakers, non-English audio, or need anything beyond “view in Notes” (DOCX, SRT, JSON export). Most of the M4A files people actually want transcribed fit one of those four criteria.

The right answer isn’t uswhen it’s a single short English voice memo and you have an iPhone with iOS 18+ — the built-in transcription is free, instant, and on-device. Don’t pay for what Apple already gives you.