Dictation vs transcription: what is the difference, and which one you need on a Mac
By Ryleigh Newman ยท Published
Dictation is speaking live and watching your words land at the cursor in whatever app you are using; transcription is turning a recording that already exists into text. Both are speech to text, and the same speech model can do either job, but they serve different moments: dictation is for writing now, and transcription is for getting the words out of an interview, a lecture or a voice memo afterward.
On a Mac, Apple Dictation does only the live job, and Voice Memos and Notes transcribe only what they record themselves. YapToText, the free app I built, does both on your Mac: one key dictates into any app, and the Transcribe a file card on the Utility page turns an audio or video file into plain text. I built it because my hands make typing difficult, and I dictate every day. If you already know what you have in front of you, skip to when you want which.
Download for free. No account, nothing locked, macOS 14 or later on Apple silicon. Source on GitHub.
On this page: 11 sections
Steps
- To dictate, click where the words should go in any app.
- Tap Right Command, talk, and tap it again to type the words at your cursor.
- To transcribe a recording, open the Utility page and choose a mode in the Transcribe a file card.
- Drop the audio or video file on the card, or click Choose File.
- Click Copy, or Insert to type the transcript into the app you were last using.
The difference in one table
The two jobs differ in where the words come from and where they end up. Dictation starts with your voice and ends at your cursor; transcription starts with a recording and ends with a transcript.
| Dictation | Transcription | |
|---|---|---|
| What goes in | Your voice, through the microphone, right now | A recording that already exists: an audio file, a video, a voice memo |
| When the text appears | As you speak, or as soon as you stop, while you are still in the app | After the file has been worked through |
| Where the words go | At the cursor, in Mail, Slack, Pages or any other app | Into a window or a card, for you to copy, insert or save |
| Who is speaking | Usually you, close to the microphone | Often several people, sometimes across a room |
| What you want back | Finished sentences, ready to send | The words, and for some jobs timestamps or who said what |
| How you check it | Read it right away, while you remember what you said | Read it against the recording, and play back the parts you doubt |
Speech to text is the umbrella term for both. Voice to text and talk to text usually mean dictation, and transcribe usually means a file, so when someone compares speech to text with transcription, they are usually asking about live dictation against a recording.
There is one more difference people miss. Dictation usually wants finished text, and transcription often wants every word as it was said. YapToText handles both with modes: Raw Transcription types exactly what was said, with basic punctuation and no AI, and Clean Up removes filler words and false starts and fixes grammar while keeping your wording. Every mode is editable, built-ins included.
Which Mac tools do which
The Mac's own apps split the two jobs between different places, and the third-party apps lean one way or the other. Ask two questions of any tool: does it type where my cursor is, and will it open a file I already have?
Checked October 3 and October 5, 2026, on support.apple.com and each maker's own pages. Features change, so check before you rely on one.
| Tool | Live dictation at your cursor | A file you already have | What a file gives you |
|---|---|---|---|
| Apple Dictation | Yes, anywhere you can type | No; it listens to the microphone | Not a Dictation job |
| Voice Memos | No | Only its own recordings, on macOS 15 or later with Apple silicon | A transcript to view, search and copy |
| Notes | Through Apple Dictation | Only audio recorded inside a note, on a Mac with M1 or later | A transcript to search and copy |
| Live Captions | No | No; captions while audio plays | Captions in a window; Apple's guide describes no transcript to save |
| YapToText | Yes, from one key, in any app | Yes, any audio or video file macOS can decode | Plain text, with no timestamps, subtitles or speaker labels |
| MacWhisper | Yes, in the version sold on macwhisper.com, not the App Store edition | Yes | Text and .srt or .vtt subtitles free; speaker recognition on Pro |
| Otter.ai | Not described on Otter's pages | Yes, imported files up to 5 GB, transcribed in Otter's cloud | Speaker labels and timestamps |
| superwhisper | Yes | On Pro only | Text shaped by your active mode |
Apple's Dictation guide describes speaking wherever you can type, and Dictation stops by itself after 30 seconds without speech. Apple's transcripts of recordings live in Voice Memos and Notes instead, and both transcribe only what they recorded. The guide to using dictation on a Mac covers Apple's live side step by step.
Among the apps, MacWhisper is a file tool that also dictates, and YapToText is a dictation tool that also does files; the MacWhisper alternative page sets the two side by side. Otter calls itself a meeting agent, and its Notetaker joins Zoom, Google Meet and Microsoft Teams calls. The best transcription apps for Mac ranks the file side in detail.
Dictate live with YapToText
Live dictation is what YapToText is for first, and it works the same way in every app. It needs an Apple silicon Mac with macOS 14 or later, and it is free on the Mac App Store. Tap, talk, tap, and the words land where your cursor is.
Step 1: Click where the words go
- You: Click into the email, the note, the document or the message box.
- YapToText: Nothing yet. It types into the app in front, never into its own window.
- Check: The cursor is blinking where you want the words.
Step 2: Tap Right Command and talk
- You: Tap Right Command once and say what you want to write.
- YapToText: Opens the floating panel with your voice as a wave and your words as they form.
- Check: The wave moves when you speak. If it stays flat, the app is not hearing your microphone yet.
Step 3: Tap it again
- You: Tap Right Command again.
- YapToText: Finishes the transcript on your Mac, pastes it at your cursor and puts back whatever you had copied before.
- Check: The words are in the app. If they went to the clipboard instead, Accessibility is not on yet.

Three things make this dictation rather than a raw transcript. Auto mode, on by default, decides how each dictation is written: a letter becomes an email, a short chat line stays as spoken, and everything else gets a careful pass that fixes what was misheard. Intelligent Insert reads the words around your cursor, so a phrase dropped into the middle of a sentence lands lowercase, with one space on each side and no stray period. And Auto-stop after silence is off by default, so a pause to think does not end the dictation. The guide to dictating into any app on a Mac goes app by app.
Transcribe a recording with YapToText
The transcription side lives on the Utility page, in a card called Transcribe a file. The same Whisper model that handles live dictation does the work, on the Mac, and the file never leaves it.
Step 1: Choose the mode before the file
- You: In the Transcribe a file card, open the mode menu and pick Raw Transcription, or a mode such as Note or Email.
- YapToText: Uses that mode for the file. Raw Transcription keeps every word, and the other modes clean it up on the way.
- Check: The menu shows the mode you want.
Step 2: Drop the file
- You: Drag an audio or video file onto the card, a Voice Memo included, or click Choose File.
- YapToText: Reads the file by its content, not its extension, and transcribes it on the Mac. A long file is cut at quiet moments and handled in pieces.
- Check: The card says it is transcribing your file. A file with no speech in it says none was found.
Step 3: Take the text
- You: Click Copy, or Insert to type the text into the app you were last using.
- YapToText: Copies or inserts plain text. Send below passes it to the next card, where Transform text can run it through another mode.
- Check: The text is where you wanted it.
Containers it recognizes include MP4, MOV, M4A, M4V, 3GP, WAV, AIFF, CAF, FLAC, MP3 and AMR, and for a video it transcribes the soundtrack. There is no minute cap and no account. Your dictionaries and spoken commands apply to files too, so the names you have taught the app come out right.

Transcribe file in the menu bar panel, and on the History page, does the same job with your current mode: it copies the text to your clipboard, types it into the app you were using if Insert text automatically is on, and files it in History under the file's name. The Utility card does none of that, and its text stays in the card until you take it. The free audio transcription guide goes further, and the Utility guide covers the other cards.
When you want which
Pick by where the audio is, smallest commitment first. If the words should land somewhere now, dictate; if the audio already exists, transcribe.
- An email, a message, a document. Dictation. Click into the app and talk; there is no file to manage.
- A thought you had away from the Mac. Record a voice memo on your phone, then transcribe it on the Mac. A voice memo is a dictation that happened somewhere else. The Voice Memos guide shows how to get the file out.
- An interview. Transcription, of the recording. YapToText gives the words as plain text, and you add the speaker names by hand; the guide to transcribing an interview on a Mac walks through it.
- Who said what, and when. Transcription with speaker labels or timestamps: MacWhisper's free version exports timed subtitles, its Pro version adds speaker recognition, and Otter labels the speakers you tag.
- A live meeting with several people. Otter's Notetaker joins the call. YapToText does not join calls or record system audio.
- A lecture you are sitting in. On a Mac with M1 or later, Notes can record into a note and transcribe it, or record now and transcribe the file later.
- A long draft with pauses to think. Dictation, in an app without a silence stop. Apple Dictation ends after 30 seconds without speech, which the 30-second limit page explains.
A dictation you get wrong costs you one sentence, which you fix on the spot. A transcript you get wrong can be quoted, filed or sent long after anyone remembers the recording, so read it against the audio before it matters.
Where the two jobs overlap
In YapToText the line between the two is thinner than the names suggest, because both run through the same pipeline. The engine is the same; what changes is where the words come from and where they go.
- Every dictation is a recording too. History keeps each dictation with its raw transcript and, by default, the recording itself, in Settings, Advanced, History & audio, under Also save the audio recording. Play it back, and you can tell whether you said it or the speech model misheard it.
- A dictation can be shaped again. Every entry keeps its raw transcript, and Regenerate in the menu bar panel runs your last dictation through any mode, so one made as a message can come back as a note.
- A crash turns into a transcription. The recording streams to a file from the first moment of audio. If the app or the Mac dies mid-sentence, the next launch transcribes the saved audio and files it in History as Recovered after a crash.
- A long dictation is handled like a file. Past 4 minutes, it is cut at quiet moments into pieces of about 2 minutes, which are transcribed as they finish, so the wait at the end covers only the last piece.
- Dictation with nowhere to go. The Dictate card on the Utility page records like any dictation, but the text lands in the card's scratchpad instead of another app.

The History guide explains what each entry keeps and for how long.
If the words come out wrong
The fixes overlap, but the first step differs: for dictation, check the microphone; for a file, check the recording. Climb this ladder, smallest step first:
- Watch the wave. When dictating, a flat wave means your voice is not reaching the app. If YapToText says "Didn't catch that," it heard silence; move closer or pick your microphone in the Input picker in the capybara's panel.
- Play the file. When transcribing, listen to the recording. A silent file, or a voice that is faint at the source, is something no model can recover.
- Look at the raw words. In History, Show pipeline details shows what Whisper heard before any cleanup. For a file, run it again with Raw Transcription.
- Set the language. The Language picker on the AI Models page starts on your Mac's system language. Choose the language spoken, or Auto-detect.
- Teach it names. Add people, products and jargon on the Dictionaries page before you dictate or transcribe; an entry shapes what Whisper hears, not just what is typed.
- Convert the file. If the card shows an error, export the audio to a common format such as M4A or WAV and drop that instead.
- Try a bigger model. Download one from the Dictation Model Library on the AI Models page. Bigger models hear better; smaller ones are lighter on battery and memory.
The guide to improving dictation accuracy goes through each fix in more depth.
What each one cannot do
- Apple Dictation. It cannot open a file, it stops after 30 seconds without speech, and the text below Dictation in Keyboard settings says whether your voice is processed on the Mac or sent to Siri servers.
- Voice Memos and Notes. They transcribe only what they record, on supported Macs, and not in every country, region or language.
- YapToText, for files. Plain text only: no timestamps, no subtitles and no speaker labels. It does not record system audio or join calls, and files transcribed on the Utility card are not filed in History.
- YapToText, anywhere. It needs an Apple silicon Mac with macOS 14 or later, and nothing is claimed for Intel Macs.
Nothing on this page ranks any tool by how well it hears. It compares what each one does, not word error rates. For files, YapToText gives you the words, not who said them or when.
The short version
- Dictation. Your voice, now, typed at your cursor.
- Transcription. A recording that already exists, turned into text.
- Apple. Dictation is live only; Voice Memos and Notes transcribe their own recordings.
- YapToText. Both, free and on your Mac, with plain text from files.
Pick by where the audio is: still in your head, dictate; already in a file, transcribe.
If something here does not match what you see on your Mac, open an issue on GitHub. Every one gets read.
Download for free. No account, nothing locked, macOS 14 or later on Apple silicon. Source on GitHub.
Related
Sources
- https://support.apple.com/guide/mac-help/use-dictation-mh40584/mac (checked 2026-10-03)
- https://support.apple.com/guide/voice-memos/view-a-transcription-of-a-recording-vm4a03609f0d/mac (checked 2026-10-03)
- https://support.apple.com/guide/notes/record-and-transcribe-audio-apdb5106e334/mac (checked 2026-10-03)
- https://support.apple.com/guide/mac-help/get-captions-of-spoken-and-computer-audio-mchldd11f4fd/mac (checked 2026-10-03)
- https://www.macwhisper.com (checked 2026-10-05)
- https://docs.macwhisper.com/article/14-how-to-use-the-dictation-feature (checked 2026-10-03)
- https://docs.macwhisper.com/article/40-macwhisper-whisper-transcription-difference (checked 2026-10-05)
- https://otter.ai/ (checked 2026-10-05)
- https://help.otter.ai/hc/en-us/articles/4425393298327-Otter-Notetaker-Overview (checked 2026-10-05)
- https://help.otter.ai/hc/en-us/articles/360047733574-Import-an-audio-or-video-file (checked 2026-10-05)
- https://help.otter.ai/hc/en-us/articles/360048465453-Adding-speakers-to-a-conversation (checked 2026-10-05)
- https://superwhisper.com (checked 2026-10-03)
- https://superwhisper.com/docs/get-started/transcribe-files (checked 2026-10-05)
- https://apps.apple.com/us/app/yaptotext/id6786382289 (checked 2026-10-03)
- https://github.com/ryleighnewman/YapToText (checked 2026-10-03)
Questions
What is the difference between dictation and transcription?
Dictation turns what you say right now into text at your cursor, in the app you are using. Transcription turns a recording that already exists into text. Both are speech to text, and the same speech model can do either.
Is dictation the same as transcription?
No. Dictation is live and lands where you are typing; transcription works on a recording after it was made. Some apps, such as YapToText and MacWhisper, do both.
Is speech to text the same as transcription?
Speech to text covers both jobs. Transcription usually means a recording you already have, and dictation usually means speaking live into an app.
Can Apple Dictation transcribe an audio file?
No. Apple Dictation listens to your microphone. Voice Memos and Notes transcribe recordings made in those apps on supported Macs, and for any other file you need an app such as YapToText or MacWhisper.
Do I need dictation or transcription for an interview?
Transcription, of the recording. YapToText gives the words as plain text on your Mac. For timestamps or speaker labels, MacWhisper's free version exports timed subtitles, its Pro version adds speaker recognition, and Otter labels speakers you tag.