End Conversation


Nuance Conversations is just about over. We had a blast and lots of people seem to be intereested in my company. I’m an idiot because I missed out on the snorkeling event earlier today. I stopped back to my room to change out of my hot sweater and by the time I got back to the resort all the boats and buses had left. No big deal, I had fun anyway. There was a boat ride last night and dinner and dancing tonight. These Nuance people really know how to party. I mean, they weren’t playing any of my favorite hits (M.O.P. Ante Up, K.R.S.-1 I’m still number 1, Kool G. Ill Street Blues) but I still enjoyed the dancing. Big up to the lady I danced with, (hope my wife’s not online reading) now there’s one other person on the planet who realizes what it’s like to dance with a giraffe. I spoke with a lot of people these past couple day regarding speech technology and learned alot about different business models emerging from the break-throughs in speech. You got Free-DA (that’s free directory assistance for my acronym challenged brethren) and mobile search dollars that are out there for the taking. And who knew speech reckognition was so advanced?

These Nuance guys, lemme tell ya’ they got the speech stuff on lock. They can recognize things that you wouldn’t believe. Imagine pulling up to your local fry/burger joint and placing and order at the drive through. Then imagine pulling around to the casheir to find out she is a Dell PC (with core duo of course) and a flat screen monitor! That’s where we’re at today. I’m mean, really think about it. What do you sound like when you place your orders? “Gimme a Big Mac… and uhh… a large fry… umm… a Dr. Pepper.”

“Ok, is there anything else I can get for you?”

“Are the quarter pounders still on sale for $0.99?”

“Yes, would you like to order one?”

“Sure, forget the Big Mac and just gimme one of those Quarter Pounders.”

“No problem! Your total comes to $2.99. Please pull around to the second window.”

That’s what these speech processors are doing today. I had a guy tell me his business sells solutions to cable providers that can completely automate the most complicated technical support calls… calls lasting over 30 minutes! Could you talk to a computer for thirty minutes? Well that’s where we are today with speech and I gotta say it is doggone impressive. Speec is one of those things where it either works or it doesn’t. And when it does work it leaves an impression that can’t be matched with any other type of interface. I’ve been using voice dialing, no I’ve been depending on voice dialing, wait a minute, I’ve been living by voice dialing on my BB 8830 for a while now. I don’t even like to use regular telephones any more because they feel like a waste of time. Who wants to punch in all those numbers?I just press the button on my bluetooth headset and say “call whomever” then the phone starts to ring at the other end. I fell in love with my Mac after I learned it could tell a joke or two. I will say this, however. Speech doesn’t work well with everything. For example, I wouldn’t write Java code with speech no matter how good the recognizer was. I wouldn’t like to flip the T.V. channels using speech and I probably wouldn’t do to good in Mortal Combat with a speech enabled controller. Still there are a lot of good uses for speech particularly with aggregating commands. Any situation where you need to walk through several menus or perform multiple contiguous tasks are candidates for a good speech recognizer. It’s late and I’m babbling about nothing. I’ll leave you with your thoughts if you leave me with a comment..

Let’s have a conversation


I forgot to tell y’all… I’m going to be at Conversations2007 this week. Well not “I’m going to be” rather I’m at Conversations2007. It’s all about voice/speech technoogy and I gotta say the stuff is way more advanced than it was a few years ago. Already I was given the tour of the Mercedes Benz speech support. It’s bangin’! I get in the front passenger seat of this car (that I initially thought was the door prize) and the guy starts talking to the steering wheel. I mean he’s not speaking directly into the steering wheel like you would have to do with 1999’s speech technology. Instead he’s having a natural conversation with the Benz. He’s saying things like “switch the station”, and the Benz responds with “what station?” Before I could blurt out “Power 99 FM my favorite station!” he cuts me off with 97.3 and some pop music ensues. (I also had to remember that I was outside of the Philly broadcasting area.) Then the guy demos the on-board dial via blue-tooth feature. He opens up a call using bluetooth to his cell phone and the call plays through the stereo speakers while the hidden in dash microphone (Where is that thing anyway I’m guessing the steering wheel) relays voice to the called party. All of the phone’s contacts can dialed this way. He finally demoes the GPS system. The whole thing was impressive.

I guess I’m used to the “If you want to pay your bill online press or say ‘one’. If you want to speak to a customer service rep press or say ‘zero’.” technology that still shows up in many customer service systems. What’s the deal with these old systems anyway? Why would I say “one” when I can press 1 much quicker? Why do companies still think this is a cool way to automate calls? I like the system my electric company has. It makes you feel like you’re talking to a real human. In fact if I didn’t know better I’d say they actually sub a human in some areas using the ol’ Mechanical Turk approach. It responds with things like, “ok, I’m not sure I quiet understood you. Did you say, ‘send the doggone repair truck to my house’?” That just feels more natural than, “your response could not be interpreted. Please press star to repeat and speak your text while holding you phone handset at the proper ninety degree angle.” I’m rambling… I’ll rap more with y’all. I’m outy 5000…
(Or is it I’m Audi 2000? Oh no I think it’s something like I’m a 97 SAAB… no wait SAABs weren’t the thing back in 97… what was that saying? I’m out like lightning? No, no, no! I’m gone Ghost like Swaze? Peace like Casper? Grrr!!! I’m losing my, my ability to swing the lingo. Yall don’t hear me tho’)