Furthermore, when you consider that MFT has a voice recognition vocabulary of 10,000 first-level commands compared to Toyota's Entune vocabulary of only 26, one would expect a higher rate of "problems" with MFTs voice control, however, the data actually shows MFT performing well compared to the industry, despite it's dramatically more complex vocabulary.
I'm about 150 pages behind in this discussion, but what the heck, I'll jump in anyway.
There's nothing wrong with MFT's voice recognition (other than being utterly archaic compared to the bar that Google sets). It works fine. That's not really surprising because Ford didn't write it. It's 3rd-party sourced from Nuance and all they had to do was interface to it and provide word lists.
When you press the command button and it does nothing, or it draws the correct screen to choose a match, but leaves it blank and tells you to choose a number -- THAT'S Ford at work.
Now, this is the first time I've ever heard of Toyota Entune, but my 3 seconds of research tells me you're way off with the 26-to-10,000 comparison. The 10,000 commands is a marketing number (please feel free to figure out how they came up with this number, I'm sure someone already has). Both systems are presumably essentially unlimited if you count POI names/media names as unique command variations.