Be ready before the turn
Cadence prepares choices while the conversation is still moving, so a thought can arrive in time to matter.
Conversation, on your terms
Be in the conversation again.
Cadence listens ahead and stages the words you may want to say, so you can respond, repair a misunderstanding, hold the floor, or start something new before the moment moves on.
Ready to say
“I love that idea, kiddo. A little fresh air sounds like the right playlist.”
One tap to speak
The room moves fast. Your voice should not have to wait.
No special app or setup for conversation partners. They just talk - Cadence listens and gets your words ready.
The problem
A conversation does not wait for a keyboard.
Typical conversation moves at 100 to 140 words per minute. Many AAC users communicate at 3 to 20 words per minute. The gap is not just speed. It is timing, personality, and presence.
The impact
A full turn in the conversation, arriving in time to matter.
Cadence is designed to protect the moments that are often lost first: a reaction, a joke, a boundary, a need, or a story someone wants to begin.
It measures participation, not only taps saved: replies spoken, topics initiated, time to response, repairs completed, rejected suggestions, and whether a reply sounds like the person.
See the research and product boundariesKeep agency with the person
AI suggests. The person chooses, edits, changes the tone, starts a topic, or says nothing at all.
Stay connected when technology fails
Needs, feelings, saved replies, a backup board, and device speech remain available when the network does not.
How it works
A faster path from thought to voice.
Listens
Keeps up with the room using live captions.
Prepares your next turn
Stages grounded replies in your voice while people are still talking.
Choose, repair, or steer
Choose a ready thought, correct the context, change the tone, or add an idea.
Speak and lead
Say what you mean, hold the floor, or start something of your own.
What Cadence gives you
Support for the whole conversation, not just one reply.
Every feature is designed to protect a practical moment: hearing what was said, finding words in time, correcting context, expressing a need, or opening a conversation yourself.
Live captions
Listen only when you choose. Cadence keeps the latest conversation visible and lets you repair a caption before it guides a reply.
Ready replies
Stages several short, distinct ways to respond, so timing does not depend on typing every word.
Your voice, wording, and context
Use your preferred name, details, writing samples, vocabulary, people, language choice, and conversation kits to guide suggestions without flattening your wording.
Lead the conversation
Start something, hold the floor, say exactly what you mean, or use a short idea to create fuller replies.
Essential communication and repair
Keep editable needs, feelings, reactions, repair phrases, saved replies, and a backup board one step away. A local help reminder can point back to an existing care plan without sending alerts.
Access and resilience
Use touch, keyboard, switch scanning, or optional local eye gaze. Offline basics and device voice stay available.
Built for access
Choose the access that fits today.
Cadence does not require one way of using it. Start with touch or a keyboard, use a switch if that is easier, or try eye-gaze focus when you are ready.
The eye-gaze option is an early local-camera beta. It only highlights a choice. The person always confirms before anything is spoken.
Touch, keyboard, or switch
Large targets, visible focus, scanning, and one clear confirmation let people choose the input that works today.
Optional eye-gaze focus
A local-camera beta can move focus between choices after calibration. Pick a steady, balanced, or fast focus speed. Looking never speaks for you: Space, Enter, or Select confirms.
Online help with offline basics
Live captions and prepared replies work when available. Needs, feelings, saved replies, the backup board, and device voice stay close at hand.
Why it's different
Participation, not just faster typing.
Cadence is built for the whole live turn: respond to what was said, repair context when it is wrong, hold space when you need time, and initiate what matters to you. The person, not the AI, chooses every word.
First minute
Start simply. Personalize later.
- 1. Open Cadence. Choose Start. Setup is optional.
- 2. Turn on Listen. Allow the microphone only when you want live captions.
- 3. Choose a reply. Tap a ready thought, or use a quick phrase when timing matters.
A short on-screen tour explains voice, personal context, wording preferences, repair phrases, scanning, and the optional eye-gaze beta. Cadence is an early communication tool, not an emergency or medical device.
Open Cadence