Jev Voice for Mac
A macOS voice and text controller that selects live UI actions with Jev.
A macOS voice and text controller that selects live UI actions with Jev.
A spoken or typed Mac command starts a bounded action chain. At each step Jev receives the original request, current Accessibility screen and action history and chooses one available action; Swift executes it and observes again. A separate model check verifies completion and requested literal values.
Choose the next operation category from the dynamically discovered action menu; a following request selects its target.
Original spoken or typed request, current screen controls, and prior actions and observations. When the discovered action menu is large, choices are operation categories or groups.
The user speaks or types a task in the floating macOS controller.
Swift reads the current Accessibility screen and available app actions.
Jev selects one live action or routes through a category/group when the menu is large.
Swift performs the choice, records the outcome and repeats within action and call caps.
A fresh observation and separate check confirm requested completion or report that the task cannot be completed.
The model chooses among currently discovered controls and installed apps, while the host owns execution, observation and completion checks. Large action menus are routed through groups so no discovered target is silently dropped.