Transcribe a voice memo on a Mac: the free way, and where it stops
By Andrey ChmerevI build Kekoso, a paid transcription app, which is a reason to be suspicious of an article about the free alternative. So the free route is written out in full and first, and every requirement is quoted from Apple's own guides. The one claim that is mine — that no Apple app registers as a handler for audio files — I checked on this machine and describe how, so you can repeat it.
If the recording is already in Voice Memos, the Mac transcribes it for free and this article can stop here. Most people who set out to transcribe voice memos need nothing beyond that, so the free route goes first, in full.
The rest is for the other case: the file that is not in Voice Memos, the language outside Apple’s ten, the transcript you need as a file rather than as text on screen.
The free way, start to finish
To transcribe a voice memo to text you need nothing beyond the Mac you recorded it on, provided it meets one requirement named further down. The same route handles anything you want to transcribe from voice memo files generally — including the ones other recorder apps leave in your Downloads folder.
Apple’s guide states it plainly: “With macOS 15 or later and a Mac with Apple silicon, speech in your audio recordings can be recognized and transcribed to text in Voice Memos.”
Open Voice Memos, click the recording, click the transcription button. The text appears where the waveform was. You can also watch it during recording — the waveform disappears and each word highlights as it is recognised.
To get the text out: “Select the text you want to copy, Control-click it, then choose Copy. You can open another document — a Mail message or text file, for example — and paste the text into it.”
Two useful details from the same page. Old recordings are covered: “If you open a recording made in macOS 14 or earlier, or a recording made on a device that doesn’t support transcription, Voice Memos transcribes the recording automatically if it includes recorded speech.” And one caveat, which is Apple’s own sentence rather than my summary of it: “Audio transcription isn’t available in all countries or regions.”
On the iPhone
Same feature, and the iPhone guide is more specific about what it needs: “Audio transcription is available on iPhone 12 or later in English (all variants), Spanish, Portuguese, Italian, French, German, Japanese, Korean, Simplified Chinese, and Traditional Chinese. It’s not available in all countries or regions.”
Tap the recording, choose View Transcript to read and select part of it, or Copy Transcript to take the whole thing. Recordings sync to the Mac over iCloud, so a memo made on the phone shows up on the laptop with its transcript.
Notes does the same job, and says the rule out loud
The Notes app transcribes audio too, and its guide states the requirement in a way the Voice Memos page does not: “Audio transcription is available on Mac computers with M1 or later when the device language is set to English, Spanish, Portuguese, Italian, French, German, Japanese, Korean, Simplified Chinese, and Traditional Chinese.”
Read that clause again, because it is the sentence that explains most “why is there no transcript button” questions.
The requirement people miss
It is not that the feature supports ten languages. It is that it is available when the device language is set to one of them.
That is a statement about your Mac, not about your recording. A Spanish interview on a machine running in Polish is not a supported combination, and nothing in the interface explains why the button is missing — because there is no button to explain.
So the free route has four gates, and a recording has to pass all of them:
| Gate | Requirement |
|---|---|
| Hardware | Apple silicon on the Mac (Notes says M1 or later); iPhone 12 or later |
| System | macOS 15 or later |
| Language | Device language set to one of ten |
| Region | “Not available in all countries or regions” |
Pass all four and you have a free transcription tool that is genuinely good. Fail one and there is no partial credit, no degraded mode, and no error message.
Where the built-in route stops
Voice memo transcribe workflows hit the same four walls, and none of them is about accuracy.
Assume you passed the gates. Four things remain outside what it does, and each has a different answer.
It transcribes recordings, not files
This is the big one, and it is a design choice rather than a limitation of the recognition. Voice Memos works on recordings that live inside Voice Memos. An MP3 in Downloads is not one of those.
You might expect a right-click and Open With to bridge that. It does not exist, and I checked rather than assumed: macOS keeps a registry of which apps can open which file types, and asking it directly (LSCopyAllRoleHandlersForContentType for MP3, M4A and WAV, on macOS 26.5.2) returns six or seven handlers per type, none of which is Voice Memos or Notes. Neither app claims a single audio type. The registry agrees with the design: these are recorders that can transcribe, not transcription tools that accept work.
There is a workaround that costs nothing: convert the file to M4A (QuickTime Player will do it — File, Export As, Audio Only) and get it into the app. It works, and it is four steps and a re-encode for every file you own.
The transcript is text on a screen, not a file
Everything Apple documents is a copy: select and copy on the Mac, Copy Transcript on the phone. Paste it somewhere and you have a paragraph.
What you do not have is a TXT file, an SRT file, a VTT file, or timestamps of any kind. If you need to jump to 14:32 in the audio to check a quote, or to caption a video, the built-in route ends here and nothing about it can be configured differently.
Ten languages
Ten is a lot more than zero and much less than the field. Whisper covers 99; Parakeet covers 25 European languages; SenseVoice covers Chinese, Cantonese, Japanese, Korean and English.
If your language is in Apple’s ten, this row does not matter to you. If it is Dutch, Polish, Russian, Hindi, Arabic, Cantonese or most of the rest of the world, it is the whole story, and no other row will change your mind.
Video is not audio
Voice Memos is a recorder. A lecture you filmed, a Zoom recording saved as MP4, a screen capture with commentary — none of these are files it opens. Extracting the audio first with ffmpeg is one command and free, and then you are back at the previous problem of getting the result into the app.
What to do about each one
| What you’re missing | Free route | Paid route |
|---|---|---|
| Transcribing a file you were sent | Convert to M4A, import to Voice Memos, or run Whisper from the terminal | Any local transcription app takes the file directly |
| A transcript as a file | Copy, paste, save it yourself | TXT, SRT and VTT export |
| Timestamps or subtitles | whisper-cli writes SRT | Subtitle export |
| A language outside the ten | Whisper from the terminal | An app that ships Whisper |
| Video | ffmpeg to extract audio, then as above | Video files directly |
| Twenty files at once | A shell loop over whisper-cli | Batch or a watched folder |
The free column is real, not a straw man. brew install whisper-cpp, download a model, and you have every row above at no cost, in exchange for a terminal and some patience. That trade is a good one for some people and a bad one for others, and the honest version of this article says so rather than pretending the terminal does not exist.
The paid column is where my own app sits — an app to transcribe voice memos it did not record, along with everything else you point it at: $29 once, recognition on the machine, files and video in, TXT, SRT or VTT out. I am not a neutral party about that, which is why it is the last column of a table rather than the point of the article.
When the built-in one is the right answer
You recorded it yourself in Voice Memos, and you want to read it. That is the design centre of the feature, it works, and paying for anything else would be silly.
You need the gist of one memo. Copy, paste, done.
You are on the phone and the recording is on the phone. The transcript is right there and syncs to the Mac by itself.
Your language is English, your Mac is Apple silicon and your work is your own voice notes. There is no gap for a third-party app to fill.
The short version
Voice Memos transcription is free and it is good, for recordings you made in Voice Memos, on Apple silicon, with the system set to one of ten languages, in most countries. Inside those four walls it is good and you should use it.
Outside them the wall is hard rather than gradual: files you were sent, video, other languages, timestamps and exports are not partially supported, they are not supported. That is the line where either the terminal or a paid app starts, and which of the two you pick is about your time, not about the recognition.
Every requirement quoted here comes from Apple’s own guides, read on 4 September 2026 and linked where it appears. The handler check was run on macOS 26.5.2 on 4 September 2026.
Questions people ask
Can Voice Memos transcribe a recording on a Mac?
Yes, and it costs nothing. Apple's guide states that with macOS 15 or later and a Mac with Apple silicon, speech in your audio recordings can be recognized and transcribed to text in Voice Memos. Open the recording, click the transcription button, and the text appears. It also transcribes older recordings automatically: Apple notes that a recording made in macOS 14 or earlier, or on a device that doesn't support transcription, is transcribed when you open it.
Why is there no transcript button in my Voice Memos?
Three reasons cover almost every case. The Mac is Intel rather than Apple silicon, which the feature requires. The system language is outside the supported set — Apple's Notes guide is explicit that transcription is available when the device language is set to one of ten languages. Or the region: Apple states plainly that audio transcription isn't available in all countries or regions.
How do I transcribe an MP3 or a file that is not in Voice Memos?
Not through Voice Memos, because it transcribes recordings that live in the app rather than files on disk. I checked the handler registry on macOS 26.5: for MP3, M4A and WAV, neither Voice Memos nor Notes appears among the apps macOS will open the file with, so there is no Open With route to it. Free options are converting to M4A and importing it, or running Whisper from the terminal; paid apps take the file directly.
Which languages does Voice Memos transcription support?
Ten. Apple's iPhone guide lists English (all variants), Spanish, Portuguese, Italian, French, German, Japanese, Korean, Simplified Chinese and Traditional Chinese, and the Notes guide on the Mac lists the same ten. Whisper covers 99 languages and Parakeet covers 25 European ones, so a language outside Apple's ten is the clearest single reason people end up with a third-party app.
Can I export a Voice Memos transcript as a file?
No. Apple's Mac guide describes selecting the text, Control-clicking it and choosing Copy, then pasting it into another document; on iPhone there is a Copy Transcript command. That is a clipboard, not an export. There is no TXT, SRT or VTT file, and no timestamps you can use anywhere else.
Can Voice Memos transcribe a video?
No. It is an audio recorder, and a video file is not something it opens. To get text out of a video for free you would extract the audio first — ffmpeg does it in one command — and even then you need the result inside Voice Memos or Notes, since neither takes a file from the Finder directly.
Is Voice Memos transcription private?
Apple's guide documents which Macs can transcribe and in which regions, but does not state where the audio is processed, so the honest answer is that the page does not say. If that matters for your recording, the test is the same for any app: turn off the network and try. What works offline was never sent anywhere.