1. STEP 01

   Hold one key and talk

   Hold Fn, say it, let go. Clean text lands at your cursor in whatever app you are already in — no window to open, nothing to paste.

   Fn hold to talk

2. STEP 02

   Tap it instead, and go hands-free

   A quick tap keeps listening after you let go. Tap again to stop and send. Esc throws the take away — nothing typed, nothing saved.

   Fn tapEsc discard

3. STEP 03

   Where the pill actually is

   Bottom centre of your screen, floating above every window — that small capsule is the whole interface. It is always there and it is never in the way. Drag it to any edge of any screen you like.

4. STEP 04

   The pill tells you everything

   Still means idle. Rippling means it is hearing you. A travelling ring means it is transcribing. A padlock means hands-free, and a red dot means a call is being recorded.

5. STEP 05

   None of this needs a window open

   The pill floats above whatever you are working in — your mail, a browser tab, a spreadsheet. There is nothing to switch to and nothing to navigate: the keys work where you already are.

6. STEP 06

   Forgot a chord? Hover the pill

   It opens the whole list — every key and what it does — plus your recent takes. Click it for the full history, drag it to move it.

7. STEP 07

   And when you do want the window

   Click the Sotto mark in your menu bar, top right of the screen. The whole app opens in one window: a sidebar of eleven rows down the left — and every section in this tour is one of those rows. It opens on Home.

8. STEP 08

   Sidebar → Modes

   Click Modes, the fifth row, and the page beside it becomes the five voices of the next five steps — one per key. That is where you rename one, rewrite what it does, or move it onto a different key.

9. STEP 09

   Fn 1 — Command

   Stop dictating and start instructing. "Draft a formal mail in Polish saying the pallets ship Thursday" — it does the thing and types the result, no preamble.

   Fn1

10. STEP 10

    Fn 2 — Email

    The same mumble comes out as a finished business email: greeting, the point first, a closing line — and your signature appended afterwards, so your phone number can never be invented.

    Fn2

11. STEP 11

    Fn 3 — Formal message, 中文

    Speak English or Polish; it arrives in Simplified Chinese, in the right register. SKUs and Latin brand names stay exactly as you said them.

    Fn3

12. STEP 12

    Fn 4 — Formal message

    A crisp, correctly-registered message to a customer, a partner or an authority — in the language you spoke. Add another language slot whenever you need one.

    Fn4

13. STEP 14

    Change your mind mid-sentence

    Already talking? Keep Fn down and tap another number. The take keeps running and simply becomes an email, or a formal note — the last number you tap before you let go is the one that wins.

    Fn held +2

14. STEP 15

    Space, mid-take, does two things

    This is going to be a long one? With Fn still held, press Space — the take locks hands-free and you can let go and keep talking. With Fn already up, Space drops the take back to plain Dictate: forget the mode, just write what I say.

    Fn+Space lockSpace back to Dictate

15. STEP 16

    Select anything on the screen

    Draw over any part of the screen and Sotto reads it — even text inside an image, a PDF or a video frame, where you could never put a cursor.

    Fn6⌥S easier to reach

16. STEP 17

    Translate what you selected

    A sticky note opens beside your cursor: the original on the left, the translation on the right, and a Copy button. It stays until you dismiss it — hover to keep it, Esc to close.

    Fn7⌥T same tool

17. STEP 18

    Wrong language? Switch it in the note

    The tag over the translation is a button. Pick another language and the same text is translated again, in place — no need to select it twice.

18. STEP 19

    ⌥1 — a screenshot, read out loud in text

    Drag a region and the words appear over it in the same note, ready to copy. This is the invoice, the error dialog, the screenshot someone sent you that you cannot select.

    ⌥1

19. STEP 20

    ⌥2 — the same screenshot, translated

    Read the region and translate it in one press. A Chinese supplier quote or a German delivery note becomes readable without leaving the window.

    ⌥2

20. STEP 21

    Fn 8 — translate whatever you copied

    Copy anything, anywhere, press this, and the translation is pasted for you. Nothing to select and nothing to open.

    Fn8

21. STEP 22

    Fn 9 — speak yours, send theirs

    Dictate in your own language and it arrives written in the other one. You set which language is yours once; Sotto works out the direction every time after that.

    Fn9

22. STEP 23

    A clipboard that holds more than one thing

    Copy, press ⌥C, and it joins a queue. Collect as many as you need, then ⌥V puts them back, oldest first. Fn , chooses between emptying the queue in one paste and walking it one item per paste.

    ⌥C collect⌥V pasteFn, all / one by one

23. STEP 24

    And the queue is never a guess

    The pill counts what is in it and says what the next paste will do. Hover it to read the queued items themselves. On the Fn row the same two tools are Fn 0 and Fn .

    Fn0 collectFn. paste

24. STEP 25

    Sidebar → Dictionary

    Six rows down. Spellings live here — the ones you add at the top, and underneath them the ones Sotto worked out for itself.

25. STEP 26

    It learns the words you actually use

    Add the brands and names it must never misspell — or say nothing: a term that shows up in three separate takes is promoted on its own and spelled your way from then on. Screenshots never teach it, because those are someone else's words.

26. STEP 27

    Replacements — the seventh row

    A find-and-replace that runs on your speech. Whenever Sotto hears the left side it types the right side, so the phrase you always fumble, or the brand nobody transcribes correctly, comes out right every single time.

27. STEP 28

    Snippets — the eighth row

    A short spoken trigger that expands into a whole block. Say "my address" on its own and the full postal address lands, formatted, with nothing retyped and nothing half-remembered.

28. STEP 29

    App tones — the ninth row

    A different register per app: casual in a chat, formal in Mail. Match on the app itself, or on the window's title — so Gmail inside a browser gets its own tone without touching how you write everywhere else.

29. STEP 30

    Sidebar → Notetaker

    Calls land in Notetaker, the third row. Every recorded call is one line — the app it came from, how long it ran, and what was pulled out of it. Open one and the next three steps are what you see.

30. STEP 31

    Your calls write their own notes

    Sotto notices when a call starts — Zoom, Teams, Meet, WhatsApp — records both sides, and afterwards hands you what was actually agreed. You press nothing. A red dot on the pill shows for as long as it runs.

31. STEP 32

    Events, tasks and notes — with the quote

    Every item carries the line that produced it, and says whether it is yours or theirs. Nothing is invented: an event whose date never resolved says "No date" rather than guessing one.

32. STEP 33

    One click into Calendar or Reminders

    A dated event offers Add to Calendar; a task offers Add reminder. Both land in the Mac's own apps, and the row then states that it went.

33. STEP 34

    Connections — the tenth row

    Where you sign Sotto in to the accounts a spoken command is allowed to act through: your Mac's Calendar and Reminders, then Google, Microsoft and Slack. One permission at a time, nothing switched on by default, and the tokens stay in your Keychain.

34. STEP 35

    Sidebar → Home

    The very first row — and the page the window opens on, so this one is usually already in front of you: every open task, and everything dated that is coming up.

35. STEP 36

    Home holds every open promise

    All the tasks from all the calls, in one list, with the call and the app they came from. Tick one and it drops into a Done group that stays folded away.

36. STEP 37

    History — the second row

    Every dictation you have ever made, newest first: what you said beside what Sotto typed, the app it went into, and the raw transcript one click away when you want to see exactly what it heard.

37. STEP 38

    Insights — the fourth row

    The running total: words dictated, your own words per minute, the typing time it has saved you, and the apps you dictate into most. It is the answer to "is this actually worth it", in numbers.

38. STEP 39

    Sidebar → Settings → Hotkeys

    Last row of the sidebar, and the page that opens runs Profile, Appearance, then Hotkeys. This is the one screen worth knowing where to find.

39. STEP 40

    Not ready to grant a permission?

    Fn needs one macOS permission before anything can read it. Until you give it, Sotto runs on the ⌥ keys instead — ⌥1 dictates, ⌥2 to ⌥6 are the five modes, ⌥7 selects, and ⌥8 · ⌥9 · ⌥0 translate. Only the clipboard queue waits for Fn. Or pick a different key entirely.

    ⌥1 dictate⌥2–6 modes⌥7–0 tools

Next up ·