How to Dictate in Joplin on Mac and iPhone

Short answer: Joplin has no dictation feature of its own. On a Mac, put the cursor in the note body and use a system-wide voice tool: hold a hotkey, speak, release, and the words appear in the Markdown editor. On iPhone, use a keyboard with a mic button inside the Joplin app.

Joplin is an open-source note-taking app that stores everything as Markdown, syncs through whichever backend you choose, and stays out of your way. That last quality is exactly why it has no dictation button. Joplin is not trying to be a platform with a feature for every input method. It is a Markdown editor with good sync, and it expects the rest of your system to handle the rest of your workflow.

Which means dictating into Joplin is entirely a question of what you put in front of it. This guide covers how to set that up on both macOS and iOS, and more usefully, how to handle the parts that are genuinely awkward: Markdown syntax spoken out loud, the difference between the two desktop editors, sync conflicts, and the material you should keep typing by hand.

Why Joplin Is an Unusually Good App to Dictate Into

Note apps vary enormously in how well they take dictated text, and the variation comes down to one thing: how much the editor interferes while you speak.

Rich editors that autocorrect, auto-format, auto-complete and reflow as you type fight anything arriving in a burst. Text lands, the editor reacts to it, and you spend the next minute undoing decisions you did not make. Joplin's Markdown editor does almost none of that. It is a plain text field with syntax highlighting. Words arrive, they sit there, and nothing rewrites them behind your back.

That makes it one of the calmer places to dictate on a Mac. The output is exactly what you said, which is the whole point. If you have dictated into heavier editors and found the experience jumpy, this will feel different immediately.

The Two Ways to Dictate on a Mac

Option 1: The built-in macOS dictation

macOS includes dictation in System Settings under Keyboard. Turn it on, pick a shortcut, put your cursor in a Joplin note and press it. Text goes in. It costs nothing and takes two minutes to set up, and for short notes it is fine.

The limitations show up quickly with longer material. Sessions can stop before you have finished a thought, punctuation is hit and miss, and there is no way to teach it the words specific to your work. If you keep a research notebook full of proper nouns, project code names or technical vocabulary, you will spend more time repairing than you saved.

Option 2: A system-wide hold-to-speak tool

The alternative is a dictation tool that lives in the menu bar and works in every app the same way. You hold a hotkey, speak, release, and the text is inserted at the cursor wherever the cursor happens to be. In Joplin that means the note body, the title field, the search box, or a tag field, depending on where you clicked last.

Voice Keyboard Pro works this way on macOS. There is no separate window to open and nothing to copy across afterwards. The relevant difference for Joplin specifically is that because it is system-wide, the exact same gesture works in Joplin, in your browser while you are researching, and in your terminal, so you are not learning a Joplin-specific habit that evaporates the moment you switch apps.

Where the Text Actually Lands

Joplin's desktop app has more distinct text fields than most note apps, and dictation goes wherever the focus is. Worth knowing before you start:

• The note body is the obvious target and the one that behaves best.
• The note title is a separate single-line field at the top. Dictating a title works, but keep it short, and be aware that pressing Return here moves focus rather than adding a line.
• The tag field accepts dictated text, though tags are usually single words you already know, so typing is faster.
• The search box takes dictation fine and is genuinely useful when you are hunting for a note by a phrase you remember saying.

The one thing to avoid is dictating without checking where the cursor is. In an app with this many fields, text arriving in the wrong one is the most common self-inflicted problem. Click into the body first, see the cursor blink, then speak.

The Markdown Problem, and the Honest Fix

Here is the part every guide to dictating in a Markdown app skips.

Markdown is punctuation used as structure. A heading is a hash and a space. A bullet is a hyphen and a space. Bold is a pair of asterisks either side of a word. None of this is speakable in any comfortable way. Saying "hash hash space What We Decided" produces a mess more often than a heading, and even when it works, you have interrupted the sentence in your head to narrate typography.

The fix is to stop trying. Dictate prose, add structure afterwards. In practice this means:

1. Speak the whole note as continuous prose, with normal spoken punctuation for sentences.
2. Stop. Read what you have.
3. Add the hashes, hyphens and asterisks by hand in a single editing pass, which takes seconds because your hands are already on the keyboard and you can see the whole structure at once.

This is faster than the alternative and it produces better notes, because you are deciding on structure after you know what the content is rather than committing to a hierarchy before you have said anything. The same principle applies to dictating bullet points and lists generally.

There is one exception worth making. If you have a structure you use constantly, a replacement rule beats both options. Voice Keyboard Pro's Smart Vocabulary is a personal dictionary with replacement rules, so you can map a spoken phrase to whatever text you want it to produce. Say a short trigger phrase, get the exact Markdown scaffold you always use for meeting notes or a daily log. You are not dictating syntax, you are dictating a name for a shape you have already defined.

Teaching It Your Own Vocabulary

The single biggest determinant of whether dictation sticks in a note app is whether it gets your specific words right. Note-taking is where personal vocabulary concentrates: people's names, project names, internal acronyms, the technical terms of whatever you happen to be deep in.

Build the dictionary from real errors rather than by guessing. Dictate normally for two days. Every time something comes out wrong in a way you have to correct, add the correction. After a couple of days you will have a short list that covers most of what actually goes wrong for you, and the accuracy jump is noticeable. Our guide on custom vocabulary that learns your words covers the method in more depth.

There is a second benefit that matters specifically in Joplin. Because Joplin's search is text search across your whole notebook, spelling consistency compounds. If a project name is transcribed three different ways across six months of notes, you will not find those notes later. Getting the term right at capture time is not cosmetic, it is what makes the archive usable.

Markdown Editor Versus Rich Text

Joplin's desktop app offers a Markdown editor with a preview pane and a rich text editor. Dictation behaves differently in each, and it is worth choosing deliberately.

The Markdown editor is the better target. Text arrives as plain characters, nothing reformats, and what you see is what gets stored. If you dictate a lot, stay here.

The rich text editor applies formatting as you go, which introduces exactly the kind of interference described earlier. It is not broken, but a long dictated passage arriving all at once is more likely to produce unexpected results, particularly inside lists where the editor is managing indentation on your behalf. If you use the rich text editor for other reasons, dictate in shorter runs and check the result.

To-Dos, Titles and the Capture Habit

Joplin distinguishes notes from to-dos, and to-dos are where dictation pays for itself fastest. The reason is timing rather than speed. A to-do you can capture in four seconds gets captured. One that requires opening a laptop, finding the notebook and typing gets remembered instead, which means it gets forgotten.

The realistic workflow is to dictate the task in the words you would use out loud, complete with the context you will need later. "Ask Priya whether the export deadline moved before I rewrite the section on data retention" is a better to-do than "export deadline" and takes three extra seconds to say and thirty extra seconds to type. That gap is the entire argument.

For note titles, the opposite advice applies. Titles are short, you often already know exactly what you want, and precision matters more than speed because titles are what you scan later. Type them.

The Web Clipper Workflow

Joplin's web clipper saves pages and selections from the browser into a notebook. Clipped material has a well-known problem: it accumulates. Six months later you have hundreds of saved articles and no memory of why any of them seemed important.

The fix is a sentence of your own at the top of every clip explaining why you saved it. Almost nobody does this, because typing it interrupts the flow of reading. Dictating it does not. Clip the page, click into the top of the note, hold your hotkey, say why this mattered, release, and carry on reading. That one habit converts a pile of links into something you can actually use, and it takes about five seconds per clip.

Code Blocks: Do Not

Joplin handles fenced code blocks well and plenty of people keep snippets in it. Do not dictate the code itself. Punctuation density, case sensitivity and exact whitespace make speech the wrong input method, and verifying dictated code takes longer than typing it.

Dictate the part around the code. The explanation of what the snippet does, the context you will need in six months, the gotcha you hit the first time you used it. That is prose, it is the part people skip, and it is the part that makes a snippet library worth having. There is a fuller treatment in dictating code comments.

Sync, Conflicts and One Habit That Prevents Them

Joplin syncs through whichever target you configured, and like any sync system it can produce conflicts when the same note is edited in two places before either edit has propagated. Joplin handles this by preserving both versions rather than losing your work, which is the right behavior, but conflicts are still a nuisance to reconcile.

Dictation makes this slightly more likely for a simple reason: it lowers the cost of capture, so you capture in more places. A thought on the phone, another at the desk, both into the same note within a few minutes.

The habit that prevents it is boring and effective. Capture new material into new notes, not into existing ones. Let sync settle. Consolidate later, on one device, deliberately. This costs you nothing, because a dictated capture is cheap enough that creating a fresh note each time is no burden, and it removes the entire category of problem.

Joplin on iPhone

Joplin has an iOS app, and phone capture is where a lot of its value lives: the note you take walking away from a conversation is the one you would otherwise lose.

Apple's built-in dictation works in the Joplin app as it does anywhere else. The alternative is a keyboard with its own mic button, which is what Voice Keyboard Pro installs on iOS. You switch to it inside Joplin, hold the mic, speak, release, and the text goes into the note. Two features are particularly relevant to note capture. Voice Edit lets you speak a correction rather than fighting the phone's text selection to fix a word in the middle of a paragraph. And two-way translation while dictating covers 24 languages, which matters if you take notes in one language and keep the notebook in another.

The iOS keyboard needs Full Access enabled to reach the network, which is a one-time step covered in our guide to enabling Full Access for an iPhone keyboard.

Meeting Notes Into Joplin

On the Mac there is a second route worth knowing about. Voice Keyboard Pro includes Meeting Mode, which handles speaker detection and produces AI notes, plus calendar meeting detection so it knows when a meeting is happening. The output is text, which means it goes into a Joplin note like anything else.

That gives you the thing Joplin cannot give you on its own: meeting notes that exist without someone typing during the meeting. The obvious caution is that recording other people is a consent question before it is a technical one, so say it out loud at the start. See meeting transcription on a Mac for how that fits together.

The Privacy Question

A lot of people choose Joplin specifically because it is open source, stores plain Markdown, and lets them pick their own sync backend with end-to-end encryption. If that describes you, you will reasonably want to know what a dictation tool does with your voice.

Here is the straight answer for Voice Keyboard Pro. Transcription runs through Voice Keyboard Pro's transcription engine, so audio is processed to produce text. The server stores only operational pings. No audio and no transcript content is retained on it. Your notes stay in Joplin, under whatever sync and encryption arrangement you already chose, and nothing about dictating into them changes that.

Being clear about the boundary: this is not the same as fully offline processing, and we would rather say so than let the distinction blur. If your threat model requires that no audio ever leaves the device, that is a legitimate constraint and this is not the tool for it. For most people keeping a working notebook, the relevant question is whether transcripts are retained anywhere, and they are not. There is more on how we think about this in private voice to text on a Mac.

A Ten-Minute Setup

1. Install a system-wide dictation tool and grant it microphone and accessibility permission.
2. Pick a hotkey your hands can reach without looking. It needs to be comfortable, because you will use it dozens of times a day.
3. Open Joplin, switch to the Markdown editor if you are not already there, and dictate a throwaway paragraph to confirm text lands where you expect.
4. Dictate normally for two days without configuring anything. Note the words that come out wrong.
5. Add those words to Smart Vocabulary. This is the step that makes the difference and it is the step people skip.
6. Add one or two replacement rules for structures you build repeatedly, if you have them.
7. Install the iOS keyboard if you capture on the phone, and enable Full Access.

Honest Limits

Three things this does not do, stated plainly.

It is not voice control of Joplin. You cannot say "new note in the Research notebook" and have it happen. Dictation puts text where the cursor is; navigating the app is still keyboard and mouse.

It does not solve Markdown syntax. The replacement-rule trick helps with structures you repeat, but there is no comfortable way to speak arbitrary Markdown, and any guide claiming otherwise has not tried it for a week.

And it does not make a first draft into a finished note. Spoken prose is looser than written prose. A dictated note usually needs a short editing pass, and the useful way to think about that pass is that it is mostly deleting. Getting more raw material down and then cutting is a different and easier problem than producing polished text at typing speed.

Frequently Asked Questions

Does Joplin have built-in dictation?

No. Joplin does not include a dictation feature on desktop or mobile. Voice input comes from the operating system or from a third-party tool, which is why the setup is the same as for any other text field on your Mac or iPhone.

Will dictated text break my Markdown?

No. Dictation inserts plain characters into the editor. What can look like breakage is dictated punctuation appearing where Markdown expects structural characters, which is why the recommendation is to speak prose and add syntax afterwards.

Can I dictate into Joplin's rich text editor?

Yes, but the Markdown editor is the better target. The rich text editor applies formatting as text arrives, which can produce surprises with long dictated passages, especially inside lists.

Does this work with Joplin's end-to-end encryption?

Yes. Encryption applies to how Joplin stores and syncs your notes, which is unaffected by how the text got into the note in the first place. Dictation is just an input method.

How does this compare to dictating in Obsidian or Logseq?

Nearly identically, because all three are Markdown editors with a plain text field. The same advice applies to each: dictate prose, add structure by hand, build a vocabulary from real errors. We have separate guides for Obsidian and Logseq covering their specific quirks.

What does it cost?

There is a free tier with daily limits, which is enough to find out whether dictated capture becomes a habit for you. Pro is $4.99 a month or $34.99 a year.

The Point of All This

Joplin is chosen by people who want their notes in a durable, portable format they control. That choice implies a long time horizon: these notes are meant to still be useful in five years.

The threat to that is not the file format. It is the notes you never took, because the thought arrived at a moment when opening a laptop and typing was too much friction for something that only felt half-important. Most of what is missing from a personal knowledge base was never lost. It was never captured.

A note format you control is worth very little if the capture step is expensive enough that you skip it.

You speak at somewhere around 130 to 150 words a minute without practicing, against a typical typing speed of about 40. That gap is not really about output volume. It is about which thoughts clear the bar to get written down at all, and lowering that bar is what changes a notebook over years.