News

DATE=8/6/98 TYPE=ENGLISH PROGRAMS FEATURE NUMBER=7-30270 TITLE=VOICE RECOGNITION SOFTWARE BYLINE=DAN NOBLE TELEPHONE=619-1014 DATELINE=WASHINGTON EDITOR=SWANEY CONTENT= (TELEPHONE QUALITY INSERTS AVAILABLE FROM AUDIO SERVICES) INTRO: TODAY, MOST PEOPLE ENTER AND RETRIEVE INFORMATION FROM THEIR COMPUTERS BY TYPING ON A KEYBOARD. BUT, IN RECENT YEARS, WE'VE SEEN THE INTRODUCTION OF WHAT'S CALLED VOICE RECOGNITION TECHNOLOGY. INSTEAD OF KEYING IN WORDS, THE COMPUTER RECOGNIZES HUMAN SPEECH. DAN NOBLE HAS DETAILED. TEXT: VOICE RECOGNITION TECHNOLOGY IS THE ABILITY OF A COMPUTER TO TAKE THE SOUND OF YOUR VOICE, DIGITIZE IT INTO A LANGUAGE OR A COMMAND THAT IT UNDERSTANDS, AND THEN BE ABLE TO RESPOND TO IT. SPEECH CAN BE TURNED INTO TEXT -- TO BE VIEWED ON A SCREEN -- OR TURNED INTO A COMMAND -- LIKE TELLING THE COMPUTER TO OPEN A FILE, OR TO DIRECT A COMPUTERIZED ELEVATOR, FOR INSTANCE, TO TAKE YOU TO THE BUILDING'S NINTH FLOOR. TAPE: CUT ONE -- VIA VOICE EXCERPT (0:07) "THANK YOU FOR HAVING ME ON VOICE OF AMERICA. THE VIAVOICE TECHNOLOGY WILL CHANGE THE WAY WE WORK AND LIVE...." TEXT: ...THIS IS WHAT VOICE RECOGNITION SOFTWARE, PRODUCED USING IBM'S RECENTLY INTRODUCED VIAVOICE (PRON: VEE-AH VOICE) 98 SYSTEM, SOUNDS LIKE... TAPE: CUT TWO -- VIA VOICE EXCERPT (0:22) "...IMAGINE SPEAKING COMMANDS TO SET THE TIME ON YOUR VCR, OR SPEAK WHAT FLOOR, OR WHAT ITEM YOU ARE LOOKING FOR, AND THE ELEVATOR TAKES YOU TO THE RIGHT FLOOR. (///OPT///) INSTEAD OF STOPPING TO ASK FOR DIRECTIONS, TELL YOUR GLOBAL NAVIGATION SYSTEM WHERE YOU WANT TO GO AND IT PROVIDES YOU WITH SPOKEN DIRECTIONS, EVERY STEP OF THE WAY. THE FUTURE NEVER SOUNDED SO GOOD." (///END OPT///) SEGUES TO: TAPE: JOE ORLANDO (0:03) "IT'S REALLY DESIGNED, FRANKLY, FOR EVERYBODY AND EVERY NEED." TEXT: THIS IS JOE ORLANDO, AN IBM MANAGER INVOLVED WITH THE DEVELOPMENT OF THIS NEW GENERATION OF VOICE RECOGNITION SOFTWARE. TAPE: CUT THREE -- JOE ORLANDO (0:23) "WHAT WE'RE EXPECTING TO SEE IN THE NEXT FIVE TO TEN YEARS IS THAT IT IS PERVASIVE EVERYWHERE: IN ELEVATORS, IN HANDHELD DEVICES, IN APPLIANCES, IN CARS. WHAT'S AVAILABLE RIGHT NOW IS DICTATION. THAT'S FOR PEOPLE WHO ARE PRODUCTIVITY-ORIENTED, WHO ARE TRYING TO DO MORE WORK, FASTER, WITHOUT HAVING TO WORK WITH THEIR KEYBOARD OR MOUSE, PEOPLE WHO HAVE PHYSICAL IMPAIRMENTS THAT MAKE IT DIFFICULT FOR THEM TO TYPE OR USE A COMPUTER." TEXT: THE VIAVOICE SYSTEM USES A MICROPHONE, ATTACHED TO A PERSONAL COMPUTER, TO ENTER VOICE COMMANDS. AGAIN, IBM'S JOE ORLANDO. TAPE: CUT FOUR -- JOE ORLANDO (0:29) "WHAT IS HAPPENS IS THAT THE MICROPHONE HAS TO BE SET UP SO THAT IT STARTS TO UNDERSTAND YOUR LANGUAGE. YOU CAN READ TWO SENTENCES AND YOU ARE ALREADY ABOUT 85% ACCURATE. WE SUGGEST THAT YOU READ SEVERAL LITTLE STORIES THAT YOU PUT IN THERE SO THAT THE SYSTEM GETS USED TO YOUR INFLECTIONS, YOUR INTONATIONS, AND EVEN YOUR ACCENT. BECAUSE, WE ALL SAY WORDS LIKE 'FAR', AND 'DAUGHTER', VERY DIFFERENTLY. AFTER IT GETS TO KNOW WHAT YOU'RE DOING, THEN WHAT HAPPENS IS YOU CAN PUT THE CURSOR SOMEWHERE IN A SPREAD SHEET, OR A PRESENTATION, OR INTO A WORD PROCESSOR AND BEGIN TO DICTATE TEXT." TEXT: THE TEXT, AS SEEN ON THE SCREEN, CAN ALSO BE TRANSLATED FROM TEXT TO SPEECH. TAPE: CUT FIVE -- JOE ORLANDO (0:09( "THAT'S THE ABILITY FOR THE SOFTWARE TO TAKE THINGS LIKE E-MAILS, OR WEB PAGES, OR OTHER DOCUMENTS, AND HAVE IT READ BACK TO YOU. I HAVE IT, FOR EXAMPLE, TO LISTEN TO MY E-MAILS WHILE I'M COMMUTING BETWEEN HERE AND HOME." TEXT: JOE ORLANDO SAYS THAT VERSIONS OF THE VIAVOICE SPEECH RECOGNITION SYSTEM ARE AVAILABLE IN LANGUAGES OTHER THAN ENGLISH. TAPE: CUT SIX -- JOE ORLANDO (0:15) "IT IS AVAILABLE IN OVER NINE DIFFERENT LANGUAGES. WE HAVE THEM IN CHINESE, WE HAVE THEM IN JAPANESE, AND WE HAVE EVEN A VERSION IN ARABIC. WE'RE WORKING ON PORTUGUESE FOR SOUTH AMERICA AND WE ARE ALSO WORKING WITH PARTNERS TO MAKE IT POSSIBLE FOR US TO DICTATE IN ENGLISH AND HAVE THAT TEXT COME UP IN SOME OTHER LANGUAGE." TEXT: COMPUTERS ARE NOW BEING USED AROUND THE WORLD. PROGRAMS LIKE IBM'S VIAVOICE PROGRAM WILL MAKE IT POSSIBLE FOR EVEN HIGH-TECH NOVICES TO BE ABLE TO COMMUNICATE BETTER USING THE NEW MEDIUM. 06-Aug-98 12:58 PM EDT (1658 UTC) NNNN Source: Voice of America .