Mercedes EQE, page-46

"given it’s a battery operated car any reductions on energy consumption… “5-10 times more efficient” hence contributing to more range on the battery…"

You are either being disingenuous or clueless.

Mercedes themselves said when discussing how Akida was used in the EQXX:

"The example in the VISION EQXX is the “Hey Mercedes” hot-word detection. Structured along neuromorphic principles, it is five to ten times more efficient than conventional voice control". (see here)

During the EQXX run, did the driver even use the hot-word detection?

That's why my original question was asking what percentage of total energy use do you think key word spotting (using conventional voice control in sleep mode) comprised while driving the EQXX? Can you answer that? Would it be meaningful if using Akida for hot-word detection helped the EQXX travel a metre more over a 1000km journey?

It is also important to realise they were comparing Akida to conventional voice control. There are now many other non-conventional chips that can do hot-word detection far more efficiently than conventional voice control.

In fact, let's see what a neuromorphic engineer and academic says when thinking about the efficiency of neuromorphic hardware compared to dedicated ML accelerators.