Skip to main content
Start a dictation with “Hey Flow” and describe what you want. SpeakoFlow writes the finished thing and pastes it, instead of typing your words.
1

Turn it on

Settings → Dictation → Generate with Flow. It is off by default.
2

Hold your Dictate shortcut and lead with the phrase

“Hey Flow, write a polite email asking to reschedule Friday’s call to Monday.”
3

Let go

The finished email is pasted where your cursor is.
The phrase only counts at the very start of a dictation. Say it anywhere else and the words paste as ordinary dictation, so Flow never fires by accident. Flow runs on your normal Dictate shortcut only. The AI cleanup shortcut keeps doing what it always does, and cleanup never runs over Flow output.

​
Settings

Three settings appear in Settings → Dictation once you turn Flow on:
Flow uses your assistant’s model and provider. Without one, your words paste as ordinary dictation and the overlay says why. Set one up on the Models page or under Providers.
Generate with Flow survives the assistant master switch. Turning the assistant off does not disable Flow, because they share a provider, not a feature gate.

​
Screen access

With Allow screen access on, Flow can take one screenshot when a command clearly refers to your screen, like “look at this and draft a reply”. The model decides whether it needs the capture; for a command it can answer without looking, it writes directly. Flow captures the monitor your mouse pointer is on, so a multi-monitor setup gets the screen you are working on. This is permissioned separately from the assistant’s screen access. Turning on screen vision for the assistant does not give Flow a camera, and vice versa. See Screen vision for the assistant’s controls.

​
Phrase matching

Matching ignores case and punctuation, and it tolerates the spellings speech-to-text actually produces. “Hey Flo”, “hey FLO”, and “Hey, Flow:” all trigger it, because a wake word should match by sound rather than by whichever spelling the model picked. The tolerance is bounded. A candidate word has to share the first letter, a similar sound, and be close in length. So “Hey follow the instructions” and “Hey floor plans are ready” do not trigger Flow. Once the phrase is heard in the live overlay, it is rewritten to match your configured spelling as you speak, so “Hey Flo,” shows up as “Hey Flow,”.

​
If nothing gets pasted

Flow is all or nothing. It pastes a complete result or it pastes nothing at all. You never get partial output, error text, or your raw spoken command. Every Flow recording is saved under the Flow filter in History, including failed and cancelled ones, so you never lose the thought.
Last modified on August 30, 2026