Limited grammar, on-device voice control

[NohWayJose]

One particular niggle I have when using OSMand is while driving. Because there is sooo much functionality packed in, it can be impossible to make key changes while driving, particularly if that change can't be accessed from the navigation menu. Examples are: zoom out (and stay zoomed out) to see a wider context, switch between the different compass/orientation modes, show the junction after the next one, toggle the junction view on/off, and a host more.

From a driver's point of view it would be great to have voice control over a few features. Having a trigger phrase and a small grammar could keep the implementation simple and 'in-device' and letting a user customize which key features they required would help simplicity. For examples...

• "OSM next compass" (toggle compass/orientation mode)
• "OSM next junction" (show the next junction/lane visual)
• "OSM following junction" (show the subsequent junction/lane visual)
• "OSM show destination" (centre the view on the destination)
• "OSM recentre" (centre the view on the current position)
• "OSM how long to destination" (enlarge time before arrival for a few seconds or voice it)
• "OSM add marker" (add a map marker at current position, with the option of a voice note"
• "OSM mark start point" (to mark the journey start point, in the event that it was 'current position' and I've already started or just finished the journey, so I can easily reverse the route)
• "OSM divert" (to reroute from the next available junction - if there's a heavy traffic notification)
• "OSM divert via " (to reroute from current position via the named road to the same destination) This uses unconstrained grammar - the road name - so may be much more difficult to do.

Let me know what you think of the idea. Is it achievable as a plugin (I'm no coder but would be happy to work with one, if you know someone) or would it be be better as native capability?

If you'd be interested in any UX/research/independent user testing, do let me know - happy to do it pro bono as time allows - I work for IBM as a design lead and obviously their work demands would be primary and I'd need to seek permission.