COMMUNICATION AID — RAPID SERIAL VISUAL PRESENTATION
Ear to Eye listens to the room and puts each word where your eyes already are — a single fixed point, paced to the speaker. No scrolling caption to chase, no line to lose. The full transcript builds alongside, so nothing said is gone.
// WHAT IT IS
Rapid serial visual presentation got a bad name from speed-reading, where comprehension collapses because you can't glance back. Ear to Eye uses it for the opposite reason. Conversation runs 130–200 words per minute — comfortably slow for this display. The win isn't speed. It's deleting the eye-tracking workload entirely, and giving the glance-back channel to a transcript instead.
Each word appears in the same place, paced to the speaker. Your eyes never move, so none of your effort goes into finding the next word.
Every sentence accumulates beside the focus window. Tap any one to replay it word by word. Missing a word is recoverable, not final.
Single words lean on working memory, which aphasia often affects. Phrase-chunk mode trades a little eye movement back for less to hold. It's a setting, not a verdict.
Freeze the stream when you need a second. Incoming words queue, and the display accelerates to catch up when you let go.
// PRIVATE BY DESIGN
Transcripts and settings stay on your phone. We run no backend and receive nothing. Speech-to-text is Apple's built-in recognition, which may process audio on Apple's servers — we say so plainly rather than burying it, and the privacy policy explains exactly what that means.