Would practical speech recognition expand the market for such software?

[Jake]

What is your guys take on Advanced speach recognition? I mean the kind that is practical. No personal training required, less than 1 in 1000 words gotten wrong, fast, natural speaking etc. Would people actually use this, would it significantly expand the market value for this software?

Currently the best is Dragon naturally speaking which gets 'supposedly' 99 in 100 words correct but some reviews say less. Also it requires speach training.

Let me know what you think

 

[dduardo]

Speach recognition is cool the first time, but it really isn't practical for everyday use. Voices get strained after constant use, fingers much less so.

 

[Wardw]

Is Anybody using Dragon, I'm looking for something that will let me dictate a book rather than write it.

 

[hitssquad]

I have used Dragon 7, though I am not using it (or any speech recognition) currently. The faster your computer is, the better it will work. (This is because the accuracy is adjastable via a slider that basically tells the program how much CPU crunching time to spend on deciphering your words. If you have a fast computer, you can of course go ahead and slide that accuracy slider all the way to the right and be able to enjoy maximum accuracy with minimal processing lag. If you have a slow computer, it might take several seconds to figure out each phrase you speak.)

Dragon 7 worked alright for me, but I grew up in America speaking with a West-coast accent. Reportedly, if you do not speak like the human voice models the speech-recognition programmers use, you may have poor results.

I recommend joining the official Dragon discussion forum. It's a great forum and they can help you there with any technical problems or training difficulties you might have.
http://support.lhsl.com/databases/dragon/sitereg.nsf