MELISSA BLOCK, host:
This is ALL THINGS CONSIDERED from NPR News. I'm Melissa Block.
ROBERT SIEGEL, host:
And I'm Robert Siegel.
Our hands are pretty busy these days, with touch screen smartphones and other mobile devices, but more and more these days we're able to do something that comes more naturally: talk to those devices, or to our personal computers.
NPR's Joshua Brockman covers consumer business and technology for npr.org. He's been trying out some of the latest voice recognition software and apps and he joins us now. Hi, Joshua.
JOSHUA BROCKMAN: Hi, Robert. Good to be here.
SIEGEL: And we're going to start off with software called Dragons. Not new. I actually tried it out a decade ago. And you're going to give me a look now at the latest version of Dragon. What's new and improved about it?
BROCKMAN: Well, Robert, I've been testing the latest version of this software, which is version 11 on my laptop. It came out this summer. And, you know, I think some of my officemates probably think I've been working a little too hard because I've been talking a lot to my
SIEGEL: Talking to yourself a lot, yeah.
BROCKMAN: Exactly. But, you know, I found that it is really helpful for sending email or for writing stories or other documents in Word. And why don't I test it out and send an email to you. I'm not going to touch anything. It's already running on the laptop. So...
SIEGEL: I'm going to vouch for you. Joshua is not using any hands right now.
BROCKMAN: So I'm just going to put on my headset and dictate something to you.
BROCKMAN: Launch Microsoft Outlook.
SIEGEL: Wait a minute, you're touching it.
(Soundbite of laughter)
BROCKMAN: Well, unfortunately, I got an error message here from Outlook. Okay. Send email to Robert Siegel.
SIEGEL: There it is. There's an open email box now.
BROCKMAN: Message from Josh.
SIEGEL: Subject line, it says message from Josh.
BROCKMAN: Go to body field. Hey Robert, comma, it's a beautiful fall day outside and I think we should go for a walk after the broadcast. Are you free, question mark? New paragraph. Joshua. Send email.
Did you get that on your BlackBerry, Robert?
SIEGEL: Joshua, comma, not yet. I'm waiting. But I'm sure it should come in momentarily, though, because I just saw it send off of your laptop.
BROCKMAN: Well, yes.
(Soundbite of laughter)
BROCKMAN: The hazards of Internet connections. Well, you know, the software, it also does allow you to control your desktop and to search the Web. And Nuance has created some noticeable improvements in accuracy since the last version.
SIEGEL: Okay. So, that's the software for a computer, Dragon. What's out there for a mobile device?
BROCKMAN: So, Nuance, the company that owns Dragon, they've also created some free apps for the iPhone, the iPad and the Blackberry. And I'll put links to these on the All Tech Considered blog. And the apps basically allow you to dictate things and paste your words into email, text, or even Twitter or Facebook.
SIEGEL: So a lot of this software both for computers and for mobile devices is coming from the same company, Nuance. Does Nuance have any competitors?
BROCKMAN: They certainly do. Google is one of the competitors. And I went to the Android store recently. It's called The Market and I downloaded one of their apps. It's called Voice Search, and I'm going to demonstrate it here with you on a Droid X phone.
BROCKMAN: Now, in order to be able to use this new feature, though, you have to be running the latest version of Android. It's called 2.2. But it allows you to do all kinds of things, from listening to music to finding a map, with something that's called voice actions. So we can give it a try.
BROCKMAN: So, Robert, I understand you were in New York recently.
SIEGEL: I was.
BROCKMAN: Was there a favorite restaurant that you ate in?
SIEGEL: Yes. I did eat at a favorite restaurant.
BROCKMAN: So, why don't you press the search button and ask it, ask the phone to call that restaurant. Say, call that restaurant in New York City.
SIEGEL: Call Nice Matin in New York City.
(Soundbite of beep)
SIEGEL: Well, now it's giving me choices. Call Melba's Restaurant. Call Trinity Church Parish. Call American Girl Place. Call Nice Mack Ten in New York. Call My Snack Time in New York City. Not quite.
BROCKMAN: Well, it is, the foreign language issue is interesting. The...
SIEGEL: It's a French restaurant, Joshua. I mean, how can you have an app that can't hear a French restaurant's name?
BROCKMAN: Well, the aspect to the phone is there is a voice search feature, and believe it or not, Robert, this is actually available for more than a dozen languages including several types - two types of Chinese, Italian, Japanese, Polish and Turkish. And then there's also a lot of versions of English. And that's for just doing a basic Google search.
SIEGEL: So I should ask you for a restaurant that has no foreign words in its name?
BROCKMAN: Well, yeah.
SIEGEL: Let's try one more.
SIEGEL: All right.
BROCKMAN: Call Blue Ribbon Sushi in New York City.
(Soundbite of beeping)
BROCKMAN: So it actually is dialing the phone number.
SIEGEL: It's so amazing. It's like it's found the 10 of diamonds right there. It's found Blue Ribbon Sushi.
BROCKMAN: So it does that by using Google Maps and leveraging that to be able to actually, you know, make the phone call. But it does work a lot better than when I first tested just the basic Google voice search maybe about a year ago. But, you know, there still are some challenges.
SIEGEL: Still haven't gotten that email from you, Joshua.
BROCKMAN: Well, I'm sorry about that. It must be the Internet connection.
SIEGEL: And it has nothing to do with the software. It's our wireless connection in the building here, in the studio, that's holding you back, I think. But as soon as I get it I'll reply and I'll tell you about going for a walk after work.
BROCKMAN: Thanks a lot, Robert.
SIEGEL: Okay. Joshua Brockman, who covers consumer business and technology for NPR.org, with a look at what's new in voice recognition.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.