The Coming Wave of Gadgets That Listen and Obey

dgan brings us a NYTimes piece about the development of speech recognition for common gadgets. Companies such as Vlingo and Yap are marketing their software to cellular carriers to give consumers a hands-free option for tasks like finding directions and text messaging. Quoting: "Vlingo's service lets people talk naturally, rather than making them use a limited number of set phrases. Dave Grannan, the company's chief executive, demonstrated the Vlingo Find application by asking his phone for a song by Mississippi John Hurt (try typing that with your thumbs), for the location of a local bakery and for a Web search for a consumer product. It was all fast and efficient. Vlingo is designed to adapt to the voice of its primary user, but I was also able to use Mr. Grannan's phone to find an address. The Find application is in the beta test phase at AT&T and Sprint. Consumers who use certain cellphones from those companies can download the application from vlingo.com."

Comments

It may finally happen.

[moogied]

5,000 years ago man relized it could not make women listen and obey. So he started a quest to make devices that could..

Is it possible that all of mankinds dreams are coming true now?!

Re:It may finally happen.

[peragrin]

nope. because we must select double delete them all.

voice recognition is no where near reliable. I laugh at my brother as he tries to use voice dial on his cell phone, it takes two or three times to get it to work. I once sneezed and it dialed my father. a good throat clearing sounds like mother. I should try farting at it some time to see who that would Dial.

Seriously try it sometime. delicately train the system for your voice, use it for a while, and then start throwing random noise at it. Or take a song which the music track is quiet enough to hear each word clearly and play that at the microphone. It should give you all the lyrics, yet they can't sort that out. The human ear can, but a computer can't yet. voice recognition is nearly useless until it can.

Re:It may finally happen.

[ScrewMaster]

The human ear can, but a computer can't yet. voice recognition is nearly useless until it can.

Voice recognition is incredibly useful in the right context. A friend of mine is an attorney who happens to be disabled. He makes great use of voice recognition on his computer, does most of his legal work with it. Is it "conversational"? No, but it serves his purposes perfectly.

So you're right, speech recognition systems aren't as generally versatile or accurate as the human brain, but they're getting better all the time. Give it ten years or so, with improved algorithms and a sixteen core processor to handle them I think we'll be interacting with computers on a much different level. Of course, by then you'll have to know Spanish or Mandarin to use one of them.

All I can think of is...

[bennomatic]

"Open the pod bay doors, HAL."

"I'm afraid I can't do that, Dave."

Re:All I can think of is...

[value_added]

"Open the pod bay doors, HAL."
"I'm afraid I can't do that, Dave."

My take on the matter is that the reason that's all you can think of is that everything else is inappropriate, inefficient or simply too goofy for consideration.

Not to anthropomorphise electronic devices (I know, they don't like it when you do that), but I think they'd prefer to be treated anonymously and respond the most basic of instructions only. And we'd prefer they remain that way, except in very limited circumstances where the device is named Lenore.

In the Star Trek movies you'll find something similar to the above, with an occasional "Tea, Early Gray, Hot" for good measure, but the rest of the time everyone is interacting with devices using ... wait for it ... keys and buttons. And this is into the technologically advanced future where most everything is a device, including crew members. Seeing Picard, for example, say "Computer, send a message to Data telling him to work on his joke-telling skills", or to use the article's example, [asking] his phone for a song by Mississippi John Hurt, would be seen by everyone as a ridiculous use of technology and dismissed as absurd.

Voice recognition, in the abstract, is fascinating and no doubt fun, but I wouldn't want to live in a Tourettes-like world where everyone is shouting out instructions to unthinking devices, let alone work in a cubicle where the next guy's phone conversation are competing with the noise of his regular work.

So past opening and closing doors, keyboards it is. Or for those unskilled in the expressive art of the command-line, a mouse or function buttons.

Fun with Gadgets

[Knave75]

User: Please connect me with Hugh Jass
Gadget: Sorry, I could not find a Hugh Jass
User: *snicker*

heard it all before

[debatem1]

I maintain great skepticism about speech recognition as an interface. It just isn't much faster than typing, even on a cell phone- and its not that it takes so much longer to get an ideal rendering, its that even a minor error in translation results in about five seconds of prompting followed by reentry. Until they can get that figured out, or get accuracy up to a point where someone unused to giving dictation can use it, its just not that great a technology.

Re:heard it all before

[mdfst13]

It just isn't much faster than typing Sure, but it's a lot safer to do while, say, driving down the road. The problem with screen output and typed input is that you have to use both eyes and hands to operate the device. By contrast, using speech input and output only requires voice and ears. Of course, there are some circumstances where the screen/type method is superior, e.g. sending emails from your blackberry during meetings. However, there are many cases where speech is superior, e.g. driving down the road (or even just walking). Viewing speech as a replacement for screen/type is over zealous. It's really more of an alternative.

It would probably help if advocates of the technology understood this. It doesn't have to be all or nothing. Two alternative solutions can add up to a more powerful solution than either would be alone.

Limited phrasebook

[name*censored*]

Limited phrasebook technology is a lot better than voice recognition technology in a lot of devices. Given that most (well, all) devices have limited functionality (not even Steve Jobs' iPod can do his taxes for him), there's very little point in giving the device the ability to understand possibly-misdirected phrases such as "Honey, have you seen the remote?". A good approach for this technology would be to limit it to understanding alternate ways of phrasing a particular command; "Device, Get Me A Beer"/"Device, Can I Have A Beer"/"I'm Really Thirsty". This way, we'd avoid misdirected speaking (the device thinking you're speaking to it instead of to another), and could also exploit the reduced set of understandable phrases to correct for people with colds/accents/quiet voices/etc, in much the same way as limited-phrasebook devices work (only with more flexibility).

Re:Limited phrasebook

[niceone]

"Honey, have you seen the remote?"

Phone: Yeah, sure, it's cute enough, but I think I can do better.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:I wonder...

[lgw]

I can't get over this "hands free text messaging" option! What engineer had the insight "we need to give customers a way to communicate over the phone just by talking"? It's a strange world.