This invention describes a system that lets you control existing phone apps with your voice or gestures, without changing the app's original code. When an app starts, the system identifies its on-screen elements like buttons and menus. It then creates custom labels or overlays for these elements, allowing you to interact with them hands-free, for example, by speaking a command that the system translates into a touch action. The system specifically uses a customizable overlay template and a label index, which can be a simple file like XML or text.
Why it matters: Filed before advanced AI models could easily interpret natural language commands and dynamically map them to arbitrary UI elements. The voice-to-UI mapping, complex in 2022, is now more achievable with LLMs.
AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.
Reinvent this with AI