Is voice control part of our mobile future?

Bob Russell · 01-27-2007, 05:16 PM

Voice recognition and control systems were frequently mentioned as the computer input of the future. It was supposed to replace keyboard input, and even was going to provide all your home controls for lights and appliances. But then we learned how hard the problem is, and how much of human listening is tied to our understanding of language and contextual information.

But with more realistic goals, speech input might have a place in our mobile future after all. Those who have used the newer versions of Dragon Naturally Speaking, for example, will have an idea of how far such programs have come. You can actually speak text and it will be recognized. Years ago, I played around with state of the art software, and it just wasn't worth using. Now it is.

With devices like the new UMPCs, we are finding that powerful devices are coming in smaller and smaller packages for computing on the go. Together with progress in voice recognition software, it is becoming evident that you might be able to control a mobile device with speech after all.

For an example of what I mean, take a look at this video demo of speech control of a Vista system. You can both see that it's not perfect, and that it has great potential. And apparently it's much better than what was included in early versions of Vista. Of special interest for mobile computing is the later portions of the video with web browsing using voice commands.

And don't forget that it doesn't have to be entirely voice controlled to be useful. We are happy to use both the D-pad control and a fingernail when using a Treo, so why not do a combination of voice and other inputs for mobile computing also? After all, you aren't always in a location where you can speak freely anyway, such as in a theater or library.

Speech control may not be like the movies in our lifetime, but it just might become a useful supplementary input method for many people in the next few years. And what voice recognition application could be easier than e-book reading? Surely any software can learn to recognize a "turn page" command!

Anchoku · 01-28-2007, 06:35 PM

I've got to have it! No more carpal tunnel!

Hm... Of course, if I use voice recognition, I'll probably need a lot more scotch and water.

Alexander Turcic · 01-29-2007, 04:52 AM

The latest C't (German computer magazine) compares Vista's built-in voice recognition to commercial alternatives... and Vista does surprisingly well! See attached comparison matrix (in German, sorry).

jmbowman · 01-29-2007, 11:25 AM

I just don't see voice recognition as being generally useful as an interface to electronic devices, although I do see the value for accessibility reasons (people who don't have free use of their hands, etc.) I suppose part of this is due to my environment...I work in an open office where we'd all go nuts if everybody talked to their computers, and I commute on a subway system where a device wouldn't be able to pick up my voice (a friend sitting next to me has trouble hearing me). I could use such an interface at home (assuming I didn't have any friends or family visiting who would get annoyed by it), but if I have access to a desk with mouse and keyboard, I don't think the voice recognition would gain me much.

I suppose it would be marginally useful for those occasions when one of my cats suddenly comes up and decides that they want to be held...might let me start some music playing or navigate some web pages while my hands are occupied with the purrball. That's really about the only use case that I can envision for myself, though...

UncleDuke · 01-29-2007, 01:56 PM

Its bad enough with all these crazies walking around talking to themselves -- a few even have bluetooth phones and claim they're talking to someone althought mostly they just talk and the other person hung up hours ago.

Voice command. He who commands loudest wins. No way.

RWood · 01-29-2007, 04:21 PM

I can just see it now Duke, the jerk at the next table shouts at his unit and mine responds to his command of "delete all!"

mogui · 01-29-2007, 10:34 PM

Quote:

Originally Posted by jmbowman

...I work in an open office where we'd all go nuts if everybody talked to their computers, and I commute on a subway system where a device wouldn't be able to pick up my voice (a friend sitting next to me has trouble hearing me)...

I am reminded of an incident in a supermarket a few years ago. A man pushing a grocery cart in the opposite direction said, "Hello." I didn't recognize him and responded with a dubious, "Hello?" Then I noticed the cell phone by his ear.

Now I live in China where everyone has a cell phone. Imagine riding the bus with people around you shouting in Chinese! It happens every day. As soon as a voice interface is available, people will find a noisy and intrusive way to abuse it.

But there is a redundancy mechanism we need to put into effect; gestures. My wife's English is imperfect. When she can't think of a word, or when I fail to understand her, she will act out the idea. You should see her doing "alligator"!

We could put the issue of potentially noisy PDA input to rest if only our pocket machines could understand our nonverbal gestures :-)

Bob Russell · 02-01-2007, 10:40 AM

Very interesting that this discussion ended up talking about the dangers of voice control.. look at this article from the BBC about Vista having a security exploit due to voice control!
Vista has speech recognition hole.

Quote:

"The exploit scenario would involve the speech recognition feature picking up commands through the microphone such as 'copy', 'delete', 'shutdown', etc. and acting on them," a Microsoft security researcher wrote on the team's official blog.

But like most exploits, there are usually significant barriers and limitations that make it less dangerous than it sounds. Still, it is intriguing and worth keeping an eye on if you plan to use the Vista voice recognition features.

nekokami · 02-01-2007, 01:29 PM

I'm holding out for direct neural input.

Seriously, I'm in the "voice control is of limited value" group. The only time I can think of when I've ever wished for it is while driving -- I'd like to be able to select music or podcasts using voice control. (It would be especially useful during pledge drive weeks on our local NPR station...)

Then again, I type about 60wpm without looking at the keyboard. My kids want dictation software so they won't have to learn to type (or spell, they think).

Bob Russell · 02-01-2007, 01:42 PM

Warning.. OT comment: You should read Encephalopath!

Quote:

Originally Posted by nekokami

I'm holding out for direct neural input.

yvanleterrible · 02-01-2007, 02:26 PM

I'd like to be a Borg!

No more fumbling with gadgets!

01-27-2007, 05:16 PM	#1
Bob Russell Recovering Gadget Addict Posts: 5,381 Karma: 676161 Join Date: May 2004 Location: Pittsburgh, PA Device: iPad	Is voice control part of our mobile future? Voice recognition and control systems were frequently mentioned as the computer input of the future. It was supposed to replace keyboard input, and even was going to provide all your home controls for lights and appliances. But then we learned how hard the problem is, and how much of human listening is tied to our understanding of language and contextual information. But with more realistic goals, speech input might have a place in our mobile future after all. Those who have used the newer versions of Dragon Naturally Speaking, for example, will have an idea of how far such programs have come. You can actually speak text and it will be recognized. Years ago, I played around with state of the art software, and it just wasn't worth using. Now it is. With devices like the new UMPCs, we are finding that powerful devices are coming in smaller and smaller packages for computing on the go. Together with progress in voice recognition software, it is becoming evident that you might be able to control a mobile device with speech after all. For an example of what I mean, take a look at this video demo of speech control of a Vista system. You can both see that it's not perfect, and that it has great potential. And apparently it's much better than what was included in early versions of Vista. Of special interest for mobile computing is the later portions of the video with web browsing using voice commands. And don't forget that it doesn't have to be entirely voice controlled to be useful. We are happy to use both the D-pad control and a fingernail when using a Treo, so why not do a combination of voice and other inputs for mobile computing also? After all, you aren't always in a location where you can speak freely anyway, such as in a theater or library. Speech control may not be like the movies in our lifetime, but it just might become a useful supplementary input method for many people in the next few years. And what voice recognition application could be easier than e-book reading? Surely any software can learn to recognize a "turn page" command!

01-29-2007, 04:52 AM	#3
Alexander Turcic Fully Converged Posts: 18,175 Karma: 14021202 Join Date: Oct 2002 Location: Switzerland Device: Too many to count here.	The latest C't (German computer magazine) compares Vista's built-in voice recognition to commercial alternatives... and Vista does surprisingly well! See attached comparison matrix (in German, sorry). Attached Thumbnails

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
iPad Enable voice control on iPad (jailbreak)	scottjl	Apple Devices	0	05-03-2010 07:17 PM
mobile phones and the future of ebooks	jseay	Writers' Corner	8	02-26-2010 03:51 PM
Free on Kindle - The Voice	koland	Deals and Resources (No Self-Promotion or Affiliate Links)	0	12-09-2009 03:06 PM
The future of mobile e-content	Alexander Turcic	News	11	10-30-2003 07:53 AM

01-28-2007, 06:35 PM	#2
Anchoku Connoisseur Posts: 54 Karma: 164 Join Date: Jan 2007	I've got to have it! No more carpal tunnel! Hm... Of course, if I use voice recognition, I'll probably need a lot more scotch and water.

01-29-2007, 11:25 AM	#4
jmbowman Junior Member Posts: 3 Karma: 10 Join Date: Jan 2007 Location: Somerville, MA Device: Sharp Zaurus SL-C3000	I just don't see voice recognition as being generally useful as an interface to electronic devices, although I do see the value for accessibility reasons (people who don't have free use of their hands, etc.) I suppose part of this is due to my environment...I work in an open office where we'd all go nuts if everybody talked to their computers, and I commute on a subway system where a device wouldn't be able to pick up my voice (a friend sitting next to me has trouble hearing me). I could use such an interface at home (assuming I didn't have any friends or family visiting who would get annoyed by it), but if I have access to a desk with mouse and keyboard, I don't think the voice recognition would gain me much. I suppose it would be marginally useful for those occasions when one of my cats suddenly comes up and decides that they want to be held...might let me start some music playing or navigate some web pages while my hands are occupied with the purrball. That's really about the only use case that I can envision for myself, though...

01-29-2007, 01:56 PM	#5
UncleDuke books & doughnuts Posts: 882 Karma: 37857 Join Date: Jan 2007 Location: usa Device: sony reader, kindle2	Its bad enough with all these crazies walking around talking to themselves -- a few even have bluetooth phones and claim they're talking to someone althought mostly they just talk and the other person hung up hours ago. Voice command. He who commands loudest wins. No way.

01-29-2007, 04:21 PM	#6
RWood Technogeezer Posts: 7,233 Karma: 1601464 Join Date: Nov 2006 Location: Virginia, USA Device: Sony PRS-500	I can just see it now Duke, the jerk at the next table shouts at his unit and mine responds to his command of "delete all!"

02-01-2007, 01:29 PM	#9
nekokami fruminous edugeek Posts: 6,745 Karma: 551260 Join Date: Oct 2006 Location: Northeast US Device: iPad, eBw 1150	I'm holding out for direct neural input. Seriously, I'm in the "voice control is of limited value" group. The only time I can think of when I've ever wished for it is while driving -- I'd like to be able to select music or podcasts using voice control. (It would be especially useful during pledge drive weeks on our local NPR station...) Then again, I type about 60wpm without looking at the keyboard. My kids want dictation software so they won't have to learn to type (or spell, they think).

02-01-2007, 02:26 PM	#11
yvanleterrible Reborn Paper User Posts: 8,616 Karma: 15446734 Join Date: May 2006 Location: Que Nada Device: iPhone8, iPad Air	I'd like to be a Borg! No more fumbling with gadgets!