Rob Chambers gives answers on the speech reco resolution issue

Rob Chambers, who works on the Microsoft Speech Recognition team at Microsoft, took notice of Craig’s blog post and my entry regarding the speech tutorial UI requiring 1024 x 768.

I really appreciate the fact that Rob is watching and listening to feedback on the blogs. It shows that Microsoft is taking in the feedback and is paying attention to the community.


Rob had this to say in response to the posts:

I understand the feedback … and it was a painful decision to have to make … but we had to go up to 1024×768 minimum resolution for a bunch of the screens in the Tutorial that Craig didn’t show you. We tried doing that in 800×600, but it just didn’t work; especially in the “Dictation” and “Working with Windows” sections of the Tutorial.

You see, in the tutorial, we try to make the system look as much like an actual running Windows Vista PC as possible, and also give the user instructions off to the right. We ask them to interactively try things (in a directed manner) so when they’re done with the tutorial, they’ll have a great idea of how to use WSR in the real world.

We may revisit this decision in the future (for Vista + 1), but for now, you’ll have to run the Tutorial on a system that supports 1024×768. However, users can still use the old style Training window on lower resolution screens.

Windows Speech Recognition itself doesn’t have a minimum resolution that’s any different than the core OS itself. You can run it on any size desktop you want.

Rob also pointed out that he is quite aware of designing tablet pc and mobile pc friendly apps, as he worked on the speech integration and inking experience for version 1 of the Tablet PC Edition. 

Although I don’t pretend to understand the complexities involved in designing the speech reco tutorial and I’m running the risk of oversimplifying the issue, I do agree with Craig on this. Couldn’t the same IF…THEN…ELSE statement that tests for the resolution and throws up the error, instead display different screen layouts and workflows, thus tailoring the experience to the user environment?

Fascinating discussion – thanks for taking part, Rob!