Archives
 
 
 
  Special
 
 
 
  About Us
 
 
 

Newsletter
Free E-mail Newsletter from BYTE.com

 
    
           
Visit the home page Browse the four-year online archive Download platform-neutral CPU/FPU benchmarks Find information for advertisers, authors, vendors, subscribers Request free information on products written about or advertised in BYTE Submit a press release, or scan recent announcements Talk with BYTE's staff and readers about products and technologies

ArticlesSpeak Naturally


July 1997 / Bits / Speak Naturally
Joe Lazzaro

For years, the computer industry and hunt-and-peck typists have awaited the day when people could use a general-purpose voice-recognition system to talk to their PCs without having to pause between words. Much to the joy of data entry operators, secretaries, people with disabilities, and busy executives, that day has apparently arrived.

NaturallySpeaking, a new program from Dragon Systems (617-965-5200; http://www.naturalspeech.com ), represents the first generation of continuous-speech dictation syste ms for Win 95 and NT. With NaturallySpeaking, the company says, you do not have to pause between words while dicta ting documents or issuing commands to your computer.

Like many voice-recognition packages, NaturallySpeaking is still speaker-dependent: It requires you to train the software to accurately recognize your voice. Officials at Dragon wouldn't reveal exactly what techniques they used to accomplish the continuous-dictation capability other than to say NaturallySpeaking uses a new speech-recognition engine to deliver improved performance.

Though NaturallySpeaking appears to have the lead in this race, Dragon's competitors say they, too, will soon have products with similar recognition capabilities. "Everybody's going to take this step soon," says Mark Flanagan, vice president and general manager at Kurzweil Applied Intelligence, another major player in the speech-recognition arena. "From what we've seen, Dragon has made a legitimate move toward continuous dictation. But they've announced essentially alpha software. How long will it take to translate to an acceptabl e product?" Dragon says the first versions of the new product will ship by the end of June, at prices starting at $695.

NaturallySpeaking requires at least a 133-MHz Pentium processor, and the program is faster on MMX machines. The software needs 32 MB of RAM under Win 95, 48 MB under NT 3.51 and 4.0, and 60 MB of free hard disk space. NaturallySpeaking also requires a standard 16-bit sound card or built-in sound system on portables. It comes bundled with a headset-style microphone. The program has a 30,000-word active vocabulary that is memory-resident and a 200,000-word backup dictionary on disk.

Having continuous recognition for general use on the Win 95 platform appears to be a first, but it should be pointed out that other continuous-dictation products for specialized use in vertical markets are already available. "IBM has had a continuous-speech product since 1996 called MedSpeak, aimed at the radiology market," says Susan Scott-Ker, a spokeswoman for IBM speech systems. "But MedSpeak's 25,0 00-word dictionary is customized for a specific application, whereas NaturallySpeaking is for daily use in a business or home environment. We're using the information gained from MedSpeak on a more general product, which will be released later this year."

IBM officials recently introduced a Chinese continuous-speech system in Beijing and Hong Kong, but the company's showing of the software was a technology demonstration only. A spokesman said IBM will announce price, shipping date, and other details later this year. Motorola says the first products based on its Chinese-language continuous-speech system may ship by the end of '97. Kurzweil officials hint that their company might offer general-purpose continuous-dictation technology by the end of the year.

If NaturallySpeaking works as Dragon claims (look for a review in an upcoming issue), it will represent an important step in making technology that's an alternative to keyboard input available to a wider audience. As computers get more powerful, m emory prices drop, and sound cards and speech-enabled applications become commonplace, voice-recognition systems will start to move into the computing mainstream.


Enunciation Still Key

screen_link (55 Kbytes)


Up to the Bits section contentsGo to previous article: Go to next article: Geek MystiqueSearchSend a comment on this articleSubscribe to BYTE or BYTE on CD-ROM  
Flexible C++
Matthew Wilson
My approach to software engineering is far more pragmatic than it is theoretical--and no language better exemplifies this than C++.

more...

BYTE Digest

BYTE Digest editors every month analyze and evaluate the best articles from Information Week, EE Times, Dr. Dobb's Journal, Network Computing, Sys Admin, and dozens of other CMP publications—bringing you critical news and information about wireless communication, computer security, software development, embedded systems, and more!

Find out more

BYTE.com Store

BYTE CD-ROM
NOW, on one CD-ROM, you can instantly access more than 8 years of BYTE.
 
The Best of BYTE Volume 1: Programming Languages
The Best of BYTE
Volume 1: Programming Languages
In this issue of Best of BYTE, we bring together some of the leading programming language designers and implementors...

Copyright © 2005 CMP Media LLC, Privacy Policy, Your California Privacy rights, Terms of Service
Site comments: webmaster@byte.com
SDMG Web Sites: BYTE.com, C/C++ Users Journal, Dr. Dobb's Journal, MSDN Magazine, New Architect, SD Expo, SD Magazine, Sys Admin, The Perl Journal, UnixReview.com, Windows Developer Network