Automatic voice search (after delay)

[hugbear]

Describe the idea (required)

The new feature would allow for the search parameters to be input through voice recognition software, as a complement and (delayed) alternative to keyboard-entered text. This has the potential to significantly reduce interaction with the device, which is especially useful while driving.
Optionally, upon finishing the voice recognition, OsmAnd would repeat the recognized text and wait for confirmation.

Tell us about the expected behaviour (required)

Upon invoking the „Search” dialog, if a configurable amount of time (seconds) has passed without keyboard input, OsmAnd would issue a short audio prompt and then it would automatically activate Voice Input to derive search terms. There should be user-configurable options to enable/disable this feature altogether, and to configure the time delay and the voice recognition engine.

Tell us about alternatives you've considered (required)

I'm currently trying to use automation software (MacroDroid) to manipulate OsmAnd's user interface to call a voice recognition engine (Futo Voice Input). So far (not very far...) it seems somewhat feasible, but even if successful it would be cumbersome to implement for users not familiar with that particular automation solution. A built-in option would make much more sense.

Context (optional)

Non-critical („nice-to-have”) feature request.