Look at who you
want to hear.
Real-time selective captioning for the deaf and hard of hearing. Point. Tap. Read only the voice that matters.
えーと、そのプロジェクトは
Wait sorry can you hear me?
先週の会議で決まった件ですが
The noise in here is crazy
+ 4 more speakers… ↑ tap a name to focus
How it works
Point your camera at the room
FocusHear uses on-device AI to detect every face the moment you open your camera. Each person is assigned a persistent speaker identity so the app always knows who is who, even as people move around the room. Nothing leaves your device during detection.
Tap the person you want to hear
A single tap selects the speaker you want to follow. The app locks onto that face and begins transcribing only their voice. Their face is highlighted in your chosen accent color. Tap the same face to pause, or tap a different face to switch instantly.
Read only their words, clearly
Background conversations and competing voices are filtered out. Only your chosen speaker's words appear as captions in real time. Captions can be shown in your own language, and a small badge flags when the speaker sounds urgent, excited, or is asking a question.
The problem
430 million people live with
disabling hearing loss worldwide.
Current caption technology transcribes everyone. It cannot choose.
In Japan alone, over 36% of the population is over 60 — the demographic most affected by age-related hearing loss. In a crowded izakaya, a school classroom, or a busy train station, every caption app in existence today delivers an overwhelming wall of overlapping text from every voice at once. FocusHear solves what no one has solved: giving deaf and hard of hearing users the ability to choose who they listen to.