Windows Phone 7 Integrates Voice; but How Good Could It Possibly Be??

My first smartphone was a Window Mobile smartphone — a Motorola MPX220. My second smartphone was also a Windows Mobile Smartphone — an iMate Jam; my third and my fourth smartphones were Windows Mobile devices as well. And then… I was done with Windows Mobile. Say what you want, but my experience with Windows Mobile was never all that good. Too often I found myself doing more to get and keep the phone up and running than I was actually using it.

BlackBerry was next, and then the iPhone came along. When Android entered the fray, I felt that Windows Mobile was heaving its last gasp. Now Microsoft desperately wants back in the game, and I think that their Windows Phone7 looks pretty good. No, I won’t be switching, but I think that it DOES look good.

There is one thing that has me scratching my head a bit, though. If you jump over to the Microsoft News Center your find a post entitled Speech, the Experience Game-Changer, and in it you’ll find this…

…Zig Serafin, general manager of the Speech Group at Microsoft, delivered a keynote address describing Microsoft’s vision for speech and natural user interfaces (NUIs). Serafin demonstrated the latest in speech recognition technology that has been designed into upcoming Microsoft products. These products promise to deliver more elegant and accessible interfaces, allowing users to utilize their voices and, in some cases, their bodies to perform actions and access information.

I could not agree more. Speech is a huge deal for me on all my computing devices. It is especially handy on my iPhone and my iPad where I find myself using Nuance’s Dragon Dictation more and more. I am totally dependent on it, and I have found how to speak to it so that my transcriptions are consistently good and only getting better. Sure the dictation software is limited on iDevices, because of Apple’s walled-garden approach, but it is awesome nonetheless. And on devices where it is deeply integrated into the OS (like the T-Mobile 3G Slide) I can only imagine the possibilities.

And that is where the head-scratching comes in.

Microsoft NEEDS a win here. They need the first Windows Phone 7 out of the gate to be AWESOME… not good, but AWESOME. Their vision of a deeply embedded voice-to-text system is a fantastic one that I share. But I’ve never seen them execute it well. Windows had speech to text for years (and still does), but it never worked nearly as well as some of the 3rd party software options. And in the last few years the leader of the field, Nuance has not only been advancing their technology in an exponential way, but they have also been on a shopping spree that has added such products and companies as MacSpeech to their portfolio. That means they are now on Macs, Windows-based PCs, iOS devices, BlackBerry and Android … and their stuff ROCKS. They GET IT, and it is only getting better and better with each update.

Why Microsoft didn’t partner with them and use their voice-to-text engine in the core of Windows Phone 7 is beyond me. It would have saved them development and given them something that was proven to work right from the start. Instead they appear to be doing their own thing. Maybe it will be great, and for their sake I hope so. But it is hard to imagine it working as well or as consistently as the products Nuance has been rolling out. In fact, watching this promo video didn’t leave me all that impressed. And if it is not as good as a Nuance product … they only have themselves to blame.

Disclaimer:  I love Nuance’s products and have Beta tested a few, but I have no connection to them outside of using their products and blogging about them. I just love products that work well, and for me, theirs do.