Dictate in any app · Notes
Joplin is open source and end-to-end encrypted, so its users think about data flow. Typally puts your words into Joplin the moment you release the key, in whichever field you left the cursor.
We make Typally, so treat this as a vendor's guide. Where Joplin has a built-in option we say so, including when it is the better choice.None. The editor is markdown.
People choose Joplin deliberately, usually after leaving something commercial over privacy. That audience will ask exactly where dictated audio goes, and deserves a straight answer.
In practice people use it for: Notes and journals, to-do lists with context, web clippings with commentary, and technical documentation.
Typally runs in the background and types into whatever window has focus, so there is nothing to install inside Joplin:
Install, sign in, and it waits in the tray.
Put the cursor in the field you are about to fill — Joplin decides where the text lands, not Typally.
Push to talk, say your piece, release to transcribe.
Cloud transcription means audio leaves your machine even though Joplin's sync is encrypted. The note is protected; the audio that became it was not.
Typally types into Joplin's markdown editor from one key. Be clear about the trade-off it involves: your stored note stays encrypted, but the audio goes to a transcription service before it becomes text. Splitting by sensitivity — most notes dictated, the few that matter typed — is the compromise most Joplin users land on.
Someone who left a commercial notes app over privacy decides which notes matter: financial and medical ones get typed, the other ninety per cent get dictated. The compromise is explicit rather than accidental.
If encryption is why you chose Joplin, keep the sensitive notes typed and dictate the rest. Most notes are not sensitive, and pretending otherwise just means fewer notes.
Cloud transcription means audio leaves your machine even though Joplin's sync is encrypted. The note is protected; the audio that became it was not.
The stored note stays encrypted. But audio is sent to a transcription service before it becomes text, so the words passed through a third party on the way in.
No. Its editor is plain markdown.
Offline speech models exist and are noticeably less accurate today. Typally transcribes in the cloud.
The free plan gives you 100 minutes a month without a card — enough to find out whether talking beats typing for you.
Also dictate in: Logseq · Standard Notes · Obsidian · see all apps