John Pollard, CEO of Jott wrote back to me yesterday regarding the processes that Jott employs behind the scenes.  He was kind enough to grant me permission to post his reply here:

Hi Nitin,
Thanks for spending some time with Jott over the holidays, and for your interest in general.

You are correct that we use a mix of human and machine technologies.   Not only would it be amazing for a machine to have such high accuracy with names, technical jargon, etc., it would be impossible to do it burdened by car noise, random accents, zero grammatical context and the typical low bandwidth cellular phone connection.   We are dead focused on making Jott immediately useful, in situations that are realistic.

The machine part gets very interesting over time, and we’re excited by the innovation that’s going to happen there.   But right now we wanted to deliver on voice-powered, hands-free, messaging and to do lists.  Lots of good stuff coming down the road too…

Best wishes and Happy New Year.

John

So, as we sort of suspected, there is some human interaction involved.  I guess the most important thing to note here is that the service is extremely easy, and the backend magic is nearly invisible to the user.  I certainly plan to follow these guys along in their development of this clean and elegant service.