Setting the basic diameter is quite easy. Since voltage is proportional to a diameter it can be converted into a flow multiplier factor. As long as diameter keeps constant it is ok, but when diameter varies it gets complicated. It would need a complicated storage of width to position and update it while filament gets used. And the longer (bowden) the more memory you should use upon it. Biggest problem are here delta printer not having any free ram already on avr boards. And every extruder would need it's own queue. So here things get complicated and without testing device it becomes a real problem to test this.
So I think I will postpone it a bit until
1. One other user comes with a solution I can add.
or 2. I get a printer having such a sensor, so I can better play around with it.
If it really gets more wide spread one of the two will normally happen.