In a message dated 10/5/98 8:37:57 PM PDT, writes: << True.. true... Well, there's a couple big problems with bringing real voice recognition to wince machines. First - most (if not all) wince processors are optimized toward power consumption over speed. Even the fabulous StrongArm isn't really capable of the kind of performance the El-Cheapo desktops can do these days. Performance will improve - greatly over the next few years - but probably not fast enough for our tastes. The second problem is that wince (ouch!) machines aren't binary compatable. So even though they claim to have some huge market share, wince machines can't exploit that market effectively by sharing software titles. And should a byte code machine magically appear that all wince devices _can_ execute the same code ... well, you just ate up all your processing power interpreting silly little instructions - nothing left for your speech recognition. I'm waiting for the day when one of the operating systems producers finally says, "Okay, we're going to do it. The machine of the star trek universe is going to appear in this reality," and sells a computer without a keyboard. It'll happen. Someday. Whether it'll sell... well.. that's a question for the marketing gurus out there. >> That would be nice. We're still trying to scare up investors to market our wearable concept...::sigh:: It's slow going, which is frustrating. Cheers, Daryl -- Subcription/unsubscription/info requests: send e-mail with subject of "subscribe", "unsubscribe", or "info" toWear-Hard Mailing List Archive (searchable): http://wearables.ml.org
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail