Tap once and speak the way you would to a person. Say So holds the raw audio and the words at the same time, so nothing has to be repeated for the machine.
Who you are writing to, which app you are in, and what you are actually trying to do. The intent gets resolved before any text is written.
False starts, filler, and circling back drop away. The tone you picked gets applied, and the sentence comes out the way you would have written it with time to spare.
Finished text goes straight into the field you were already in. Nothing to launch, nothing to paste, no window between the thought and the send button.
“hi sry im late traffic is insane omw”
Raw speech, held on device
Resolved before a word is written
hi sry im late traffic is insane omw
Hey! Sorry, traffic is insane. Running about 20 minutes behind, on my way now.
Delivered
“tell sarah im not gonna make the 3pm”
Hi Sarah, I won’t be able to make the 3pm.
“same thing but for my manager”
I am no longer able to join the 3pm.
“so um i think we should uh push the launch”
I think we should push the launch.
“Per my previous email, kindly advise.”
Following up on my last note.
“heads up the payments fix is going out”
Heads up: deploying the payments fix now.
“diles que llego en veinte minutos”
I’ll be there in about twenty minutes.
Say So is a system keyboard, so it shows up wherever iOS shows a text field. The tone you pick in a work thread does not follow you into a text to your sister, and the context you were writing in never leaves the device without you knowing.
“hi sry im late traffic is insane omw”
Spoken into iMessage, 9:41am
We are building so that as much of the intelligence as possible runs on your phone. When we ship, we will say plainly which parts run where.
Plain-language policies covering what we collect, when, and why. No fine print that needs a law degree to parse.
Your words are not inventory. We will not sell your data, and we will not build an ad business on top of what you type.