Ryan's Voice — Conversation-Aware AAC App

The problem

Communication shouldn't take 90 seconds

Research shows AAC users achieve 4-15 words per minute through direct selection (Beukelman & Light, 2020). Ryan, our founding user, types at 4 WPM with one-finger typing. A simple 6-word sentence takes him over 90 seconds. Conversations don't wait that long.

4-15 WPM

AAC direct selection rate

Published research, Beukelman & Light, 2020. Ryan types at 4 WPM.

90+ sec

Per sentence at 4 WPM

A 6-word sentence at 4 WPM = 90 seconds. Conversations move at 2-5 seconds per exchange.

$299

What leading AAC apps charge

Ryan's Voice free tier is a complete AAC app.

Ryan stopped typing. Not because he had nothing to say — because nobody was listening by the time he finished.

Why current AAC fails

We lived every one of these problems

Ryan's Voice wasn't designed in a lab. It was built by a father who spent years watching his son struggle with tools that weren't built for him. Every feature exists because a real problem demanded it.

Pain Point

Grid layouts punish motor disorders

Apraxia causes past-pointing. Small grid buttons create cascading miss-backspace-retry cycles.

Our Answer

Large tap tiles — 60pt minimum, 90–200pt in practice, way above Apple's 44pt standard. One tap speaks. Swipe horizontally to change categories (Ryan's proven motor strength — he navigates YouTube by swiping). Low-density tile layout, not a dense symbol matrix.

Pain Point

No app listens to the conversation

Static phrase grids show the same options whether discussing food or feelings.

Our Answer

Patent-pending conversation-aware AI listens to the partner's speech, detects the topic in real-time, and surfaces relevant phrases. When the conversation shifts from school to food, the suggestions shift with it. Automatically.

Pain Point

Vocabulary insults intelligence

Simplified labels: "happy", "sad", "want food." Assumes nonverbal = cognitively impaired.

Our Answer

1,120+ phrases sourced directly from occupational therapy sessions, with 100% coverage of the Universal Core 36 (UNC Chapel Hill Center for Literacy and Disability Studies) and the Beukelman et al. (1989) preschool frequency vocabulary. College-level vocabulary: "She seems disappointed by the outcome." "He's totally lacking self-control." Every phrase matches demonstrated ability, not assumed limitation.

Pain Point

Cloud dependency breaks trust

Conversations sent to remote servers. No internet = no intelligence. Privacy is an afterthought.

Our Answer

100% on-device AI pipeline. Speech recognition, semantic analysis, topic detection, phrase ranking — all on the phone. Zero data sent anywhere. Works in the middle of the ocean. Conversations between parent and child are too intimate to leave the device.

Pain Point

Communication is paywalled

$199-$299 apps. "Free" tiers are stripped demos. Families already under pressure.

Our Answer

Free tier is a complete AAC app. 1,120+ phrases, text-to-speech, favourites, Quick SOS. Permanent. Not a trial. The paid tiers unlock AI intelligence — but nobody is denied communication because they can't afford a subscription.

Pain Point

No path beyond basic needs

AAC stops at "I need help." No tool helps users write stories, express creativity, or pursue ambitions.

Our Answer

Author Mode (coming soon) — AI writing tools that learn the user's voice. Sentence completion, word banks, story structure scaffolding, export to PDF. Ryan wants to be a published author. We're building the tools to make that possible.

How it works

Three steps. Two seconds.

Ryan's Voice replaces 90-second typing with 2-second tap-to-speak.

Choose a context

Tap a large context tile: conversation, school, home, gaming, or Quick SOS. Each loads relevant phrases instantly.

Tap a tile to speak

Large tiles with adult vocabulary — 90–200pt, well above Apple's 44pt minimum. One tap speaks. Swipe horizontally to change categories. Haptic feedback on every interaction. Built for motor strengths.

Tap to speak

One tap. The phone speaks the phrase aloud. Instant. Natural. Done in 2 seconds, not 90.

90 seconds with typing 2 seconds with Ryan's Voice

Intelligence

The first AAC app that follows the conversation

Most AAC apps show the same static phrases regardless of context. Ryan's Voice listens and adapts in real-time.

How the AI works

With Smart Voice, the app listens to your conversation partner's speech. When someone mentions food, food-related phrases appear. When the topic shifts to school, school responses surface. Automatically.

The entire AI pipeline runs on the device. Speech recognition, semantic analysis, topic detection, phrase ranking, and suggestion generation. No server calls. No API costs. No internet required.

Your conversations stay completely private. Zero data is collected. Zero data leaves the phone. Ever.

100% on-device Works offline Zero data collection Patent pending

MIC

Partner speaks

"Did you enjoy the movie?"

↓ on-device processing

AI detects topic

Entertainment + Question detected

↓ instant suggestions

Relevant phrases appear

"That was awesome!" "I want to watch it again"

Privacy

Your conversations never leave the device. Ever.

Ryan's Voice uses the microphone to listen to conversations. Here's exactly what happens with that data: nothing leaves the phone.

✓ No internet required

The entire AI pipeline — speech recognition, semantic analysis, topic detection, phrase ranking — runs on the device. In the middle of the ocean, in a desert, in a hospital with no WiFi. If there's no signal, Ryan's Voice still works. Your voice should never depend on a connection.

✓ Zero data collection

No audio is recorded. No transcripts are stored remotely. No usage data is sent to any server. No analytics. No tracking. No telemetry. Zero.

✓ No account required

Download and use immediately. No sign-up. No email. No personal information collected. The app doesn't even know your name.

✓ No cloud dependency

Other AAC apps send speech to cloud servers for processing. Ryan's Voice sends nothing anywhere. If our company disappeared tomorrow, the app would keep working exactly the same.

We believe communication is too intimate to leave the device. Conversations between a parent and child, a therapist and client, a teacher and student — these deserve absolute privacy. That's why every piece of intelligence in Ryan's Voice runs locally, with no exceptions.

Pricing

Core communication is free. Intelligence is the upgrade.

The free tier is a complete AAC app — not a trial, not a demo. The paid tiers unlock the AI conversation engine.

Free

Forever

A complete AAC app. Not a trial. Not a demo.

✓ 1,120+ phrases across 27 topics
✓ 100% research-backed core vocabulary
✓ Text-to-speech
✓ Large tap-tile interface with swipe-between-categories
✓ Favourites and custom text
✓ Quick SOS for emergencies
✓ Keyword-based suggestions
— AI conversation listener
— Topic detection
— Conversation memory
— DreamBot phone calls
— Author Mode

Join Waitlist

Designed around motor strengths, not weaknesses

Every interaction in Ryan's Voice is built for people with motor planning disorders.

✋ 60pt+ touch targets

Apple recommends 44pt minimum. We use 60pt or larger. Primary actions are 100pt+. Generous hit zones extend beyond visible boundaries.

👈 Swipe between categories

Horizontal swipe on the tile grid cycles through categories. A continuous fluid motion leverages the motor control apraxia users often retain, while single taps stay available for precise phrase selection. The best of both.

📳 Low-density tile layout

Typical AAC shows 20–60 symbols per screen. We show 4–8 large tiles with generous spacing. Conversation-aware AI surfaces the right phrases for the moment, so Ryan rarely needs to navigate deep.

💫 Haptic feedback

Every successful tap, swipe, and speak action is confirmed with haptic feedback. The body knows it worked before the eyes confirm it.

"She seems disappointed by the outcome."

An actual phrase from Ryan's therapy sessions. College-level vocabulary. Adult analytical depth. This is the level Ryan's Voice is built for.

The story

Built by a father for his son

Ryan is 16, nonverbal, autistic, with apraxia of speech — a motor planning disorder that means his brain knows what to say but can't coordinate the muscles to say it. His mind works at college level. Since he was 11, we've been capturing his vocabulary from therapy sessions — phrases like "being in awe of someone" and "totally lacking self-control" that reveal the depth of his thinking.

Every AAC app we tried either dumbed down his vocabulary or required precise grid tapping. Both wrong for a brilliant mind with a motor planning disorder. So we built something different: five years of Ryan's real words, powered by on-device AI, in an app that respects his intelligence.

The free tier is complete because core communication should never be paywalled. The AI layer is the upgrade because it genuinely changes what's possible in a conversation.

Ryan wants to be a published author. Author Mode is coming to help make that happen.

Roadmap

Where we're going

Ryan's Voice started as a way to keep up with conversations. But communication is only the beginning. Here's the journey ahead.

Now

Ryan joins conversations again

After years of frustration, Ryan has a way back into real-time conversations. Tap large tiles sourced from his actual therapy sessions, swipe to change categories, and be heard in seconds instead of minutes.

1,120+ phrases 27 conversation topics AI conversation listener Cross-session memory Quick SOS

Coming next

The app learns who Ryan is

Ryan uses certain phrases at certain times in certain contexts. The app will learn those patterns and adapt. His most-used phrases rise to the top. Morning phrases appear in the morning. The more he uses it, the more it feels like his own.

Adaptive card ranking Usage analytics Custom phrase promotion Personal Voice integration

Planned

Ryan becomes an author

Ryan has stories in his head. Characters with arcs. Plots with moral complexity. At 4 words per minute, a novel is impossible. Author Mode changes that — AI writing tools that learn his unique voice, help him structure stories, and export them to the world. Target: 500 words in 40 minutes.

AI writing assistant Story structure tools Word banks Export to PDF

Future

Every voice, every language

Ryan's Voice was built for one person first. But the architecture generalises. Multilingual support so families can communicate in their own language. iPad-optimised layouts for classroom and home use. Assistive Access integration for simplified interfaces. The goal: no nonverbal person is denied communication due to cost, language, or location.

Multilingual support iPad optimisation Assistive Access Community phrase packs

Your child is brilliant. Their AAC should reflect that.

Communication shouldn't take 90 seconds

We lived every one of these problems

Three steps. Two seconds.

Choose a context

Tap a tile to speak

Tap to speak

The first AAC app that follows the conversation

How the AI works

Your conversations never leave the device. Ever.

✓ No internet required

✓ Zero data collection

✓ No account required

✓ No cloud dependency

Core communication is free. Intelligence is the upgrade.

Designed around motor strengths, not weaknesses

✋ 60pt+ touch targets

👈 Swipe between categories

📳 Low-density tile layout

💫 Haptic feedback

Built by a father for his son

Where we're going

Ryan joins conversations again

The app learns who Ryan is

Ryan becomes an author

Every voice, every language

Be the first to know when Ryan's Voice launches