The Visual Lounge lets you find out about TechSmith behind-the-scenes. Watch screencasts and videos from other customers, meet up with your fellow TechSmith users and staff, and get more tips and tricks!

Some folks on Chris's blog complained that the speech recog was "slow." I'd guess there's a good chance that running Camtasia's Recorder at the same time as the speech recog might have slowed the recognition. Do you think Chris could comment about that?

For speech recognition, seeing (and hearing!) is an important part of believing. What I see with my own eyes is a tad less vulnerable to PR hype. Yeah, he might have practiced some things, and might have avoided some known (to him!) weaknesses, but it's a pretty nice demo. Speech recognition is a complex area, but at least Vista is in the game.

Hi Mike - Chris responded to some of the questions on his blog about the screencast and the slowness issues...check it out here: http://blogs.technet.com/chenley/archive/2006/03/17/422376.aspx

Betsy Weber

Looks like I did the right thing in switching over to the iMac Core Duo. This Vista ain't nuthin'

Chris,
What about words you dictate that the voice recognition does not know how to spell? Does it prompt you for correct spelling or can you fix by highlighting the word and telling the computer how to spell it?

It seems as if Vista offers nothing exceptionally revolutionary in the terms of speech recognition due to the performance of its 3rd party predicesors. All things that were displayed in this demo, including the, "show numbers" option is genuinely offered in software such as Dragon's Naturally Speaking software. Compared to Dragon's software it seems that vista also lacks some of the features of the third party package. For example, as in just saying right off the bat, anywhere you are "Start Internet Explorer", or any other program, without waiting for the start menu to open and it to display the program's icon.

I think the only thing at this point that would be anywhere close to revolutionary in terms of speech recognition is possibly designation distinguishment (where your voice was directed to, and distinguishment of your voice against other background noises.) and voice feedback. For example, asking your computer a question, whether common or not, and it being able to respond accordingly bypassing other applications. E.g. Ask it, "What's the weather like today?", a response, "Currently it's 68 degrees and partly cloudy, later today we may have light showers and possible thunderstorms tonight", you want to see this info so you say, "Show me" and it could bring up the corresponding information. or you could move onto a diffrent subject. Or even for example, imagine your computer announcing the arrival of a new email, stating the author, and asking if you would like it to be read to you, all while you work on your paperwork at your desk and requiring no additional attention from you. Also, imagine being able to interupt it by voice while its speaking, or for the system to recognize that you are engaged in a converstion with someone and to refrain from speaking. These are the kinds of revolutionary things I expect from voice recognition. I am not impressed Vista.

It seems as if Vista offers nothing exceptionally revolutionary in the terms of speech recognition due to the performance of its 3rd party predicesors. All things that were displayed in this demo, including the, "show numbers" option is genuinely offered in software such as Dragon's Naturally Speaking software. Compared to Dragon's software it seems that vista also lacks some of the features of the third party package. For example, as in just saying right off the bat, anywhere you are "Start Internet Explorer", or any other program, without waiting for the start menu to open and it to display the program's icon.

I think the only thing at this point that would be anywhere close to revolutionary in terms of speech recognition is possibly designation distinguishment (where your voice was directed to, and distinguishment of your voice against other background noises.) and voice feedback. For example, asking your computer a question, whether common or not, and it being able to respond accordingly bypassing other applications. E.g. Ask it, "What's the weather like today?", a response, "Currently it's 68 degrees and partly cloudy, later today we may have light showers and possible thunderstorms tonight", you want to see this info so you say, "Show me" and it could bring up the corresponding information. or you could move onto a diffrent subject. Or even for example, imagine your computer announcing the arrival of a new email, stating the author, and asking if you would like it to be read to you, all while you work on your paperwork at your desk and requiring no additional attention from you. Also, imagine being able to interupt it by voice while its speaking, or for the system to recognize that you are engaged in a converstion with someone and to refrain from speaking. These are the kinds of revolutionary things I expect from voice recognition. I am not impressed Vista.