COMPUTER SPEECH TECHNOLOGY
Recognition, Command, Control, And Dictation

A BRIEF OVERVIEW:
Computing Without Keyboard Or Mouse
HAVE YOU MET A DRAGON TODAY?

Text By: SuperAdaptoid

IN THE BEGINNING

First there was silence. Then there was speech. Some think things have gone downhill ever since. While that was long ago, and far away; in computing it was only yesterday.

Thus begins a Techno -Fairy- Tale for you in your time and place. We will speak of Dragons. Their power of language, voice and speech recognition. The consequence of unbounded growth, the thirst for power, great hunger, and a voracious appetite. These are part of the Mystery and Magic -Of- Technology.

Like all good fairy tales, this one too should start with the same disclaimer. All good and truthful stories start with "Once Upon A Time".

IN THE MAGIC -OF- TECHNOLOGY ARE MANY DRAGONS:

  • There are Dragons that hunger for power.
  • There are Dragons that listen.
  • There are Dragons that speak.
  • There are Dragons you can control.
  • There are Dragons you can command.
  • There are Dragons you can dictate.
  • There are Dragons from Dragon Systems.
  • There are Dragons from other sources and places.
These are the Dragons -Of- Technology.

By the early 1980s, there were Computers, already with Speech and Adaptive Technology. Before XTs, ATs, and the PC's Clones. Before Microsoft Dos and Windows. Before "Big Blue's" [IBM] OS2s, as well as all the Apple - Macs. There was speech and the Commodore 64. It was not "pretty", but it could speak! Only a matter of time and technology, "Before Communication Would Go Both Ways".

THE CARE AND FEEDING OF THE TECHNO-DRAGON

Early computing was slow, options were limited, and technology was archaic. Because of low powered computers and high power system demands, most Adaptive Applications were independently powered, stands- alone "smart boxes", which were ported or carded into the computer. This trend was to change as computers evolved to become bigger, better, and faster. Adaptive Applications moved out of their hardware and into the computer software. Soon we may need to "Put Power Hungry Applications On A Diet, Or Back In A Box".

The question for today is NOT whether computers will keep up with the leaps and bounds of technology. The question for tomorrow is will the users keep up the payments for boundless technology. In the meantime, between the two, "We Feed The Dragon".

LIVING WITH THE PC DRAGON:
First There Was Dragon Systems. Then There Was Dragon Dictate

Speech Recognition, Command, Control, and Dictation was soon to follow. Here is a short chronology of how it happened in the American PC Market. While Dragon Systems was number one, "Other Dragons Were Soon To Come".

  • 1982 ... Dragon Dictate Starts.

  • 1984 ... Dragon Dictate introduces its' first speech recognition product incorporated in a portable/desktop PC.

  • 1985 ... Several [smaller] software development companies enter the market with voice activated command and control, data entry and retrieval, text systems for PC's. Some are "ported", and will become [sound] card based. Others will develop their own sound cards.

  • 1986 ... Dragon Dictate introduces its' first 1000 word application for building voice activated command and control, data entry/retrieval, and limited domain text systems for PC programs.

  • 1987 ... Dragon Dictate introduces its' first DSP- based speech recognition tools with flexible application- specific acoustic and language modeling capabilities.

  • 1990 ... The first 30,000 word speech recognition system. The first commercially available large vocabulary speech to text system for general purpose dictation running on PC's. The introduction of commercial automated speech recognition, dictation, and information management systems. The introduction of large technical and professional vocabulary speech to text system for dictation running on PC's.

  • 1991 ... IBM announces Voice Type Version 1.0. A 7,000 word speech to text system, under license from Dragon Systems.

  • 1992 ... Dragon Systems, UK, Ltd. research subsidiary is founded in Cheltenham, England to focus on telephone and robust speech recognition technology.

  • 1993 ... IBM introduces Personal Dictation System [Voice Type 2] a 7,000 word speech to text system. As well as the Voice Type Control, a voice command control program for Microsoft Windows licensed by IBM Voice Pilot, based on Dragon Systems’ speech technology. Included with Microsoft’s Windows Sound System. Lanier Worldwide, Inc., introduces EMstation, an automated speech recognition dictation and information management system, based on Dragon System’s technology, for emergency medical reporting.

  • 1993 ... Dragon Voice Tools, introduces an extensive software developer’s kit and API [Application Programmer’s Interface]. Enabling programmers to integrate broad discrete/continuous speech recognition capabilities into their PC applications. Dragon Dictate 30,000 word Version 2.0 features faster and more accurate recognition, more sophisticated language modeling, larger backup dictionary with acoustic modules, completely flexible active vocabulary, automatic learning capabilities, and other improvements. Talk -To- Plus, software for control of Windows software programs, licensed to IBM for shipping with M-Wave products. ExecuVoice version licensed to Media Vision and is included as part of Pro Audio Studio 16. Japanese version is included with Epson products. Dragon Dictate introduces a 30,000 word speech to text system in German, and distributes to Germany, Switzerland and Austria.

  • Voice Pilot from Microsoft and IBM Voice Type 3, goes international with language versions in French and German. The control component of VoiceType Dictation works with continuous spoken commands and macros. Words run together as you would normally talk. While the dictation component works with discrete, or isolated, speech. A more distinct way of speaking that leaves a one- tenth of a second gap between each word.

  • 1994 ... Several small software development companies have left the market. Some have been unable to survive the first Microsoft Dos to Windows transition. Others have re directed their efforts from personal/home to commercial/industrial applications. Most are now Creative Labs Sound Blaster [sound card] compatible.

  • Covox, makes the transition from MS Dos to MS Windows 3.+ , and moves from "ported", to develop their own [sound] cards Sound Master and Voice Blaster. Successfully developing a single sound card that emulates a wide variety of other popular commercial sound cards; including many of Creative Lab's Sound Blaster cards. But Covox looses in court against Creative Lab's. Sound Master and Voice Blaster get Sound Blasted. Sound Blaster and Creative Lab's rule.

  • 1994 Dragon Dictate's first enabled dictation on industry standard 16-bit sound cards with the release of DragonDictate 1.0 for Windows.

  • 1995 ... Dragon Dictate 2.0 for Windows adds Windows 95 support while increasing dictation speed, improving new user accuracy and incorporating continuous speech for numbers and commands. Improvements in performance allow for faster and more natural PC speech dictation.

  • 1995 ... IBM Introduces Voice Type3.

  • 1996 ... DragonDictate 2.5 for Windows is the first dictation speech recognition product for Windows NT and the first to offer text to speech capabilities. The two-way speech capabilities are a result of a partnership with Centigram Communications Corporation, based in San Jose, CA. Version 2.5 also adds built in Netscape support, allowing users to access browser menus and dialog boxes by voice.

  • DragonXTools visual controls combine the compatibility, performance, speed, and accuracy improvements of Dragon Dictate for Windows 2.5 with the simplicity of industry-standard VBX components. Using DragonXTools, developers can easily control Dragon Dictate for Windows’ recognition, user interface, grammars, and even vocabulary through a simple set of properties, methods, and events from within a variety of popular visual development environments.

  • DragonLaw for Windows 1.0 streamlines legal dictation and research by adding WESTLAW and LEXIS-NEXIS speech commands. The add on language module enhances the capabilities of Dragon Dictate for Windows, offering the most comprehensive legal- specific vocabulary available, including over 1,000 predefined voice commands for the popular WESTLAW and LEXIS-NEXIS research services.

  • DragonEXTRA for windows 1.0 customizes Dragon Dictate for journalists, writers and others who use exceptionally broad, contemporary vocabularies across diverse topics. Dragon Tech for Windows 1.0 adds a comprehensive vocabulary of computer- related terms to Dragon Dictate, addressing the unique needs to computer industry professionals.

    1982 To Date
    Quick Summary

  • First they were slow, limited and archaic.
    Then there was recognition and command.

  • Then there was control and dictation.
    First they were slow, limited and archaic.

  • First there was MS Dos.
    Then there was MS Dos -and- Windows 3.+.
    Now there is MS Windows 95+ and beyond.

  • First they were ....... dumb.
    Then they were .... trainable.
    Now they are ... self- trained.

  • First speech was Discrete
    [word -by- pause -by- word].
    Then speech was Continuous
    [word -or- phrase -by- pause].
    Now speech is Natural
    [phrase -or- thought; without pause].
    Tomorrow speech may be Neural?

    • AHEAD OF THE DRAGON

    Obviously, Dragon Dictate was the head of the game. Little wonder both IBM and Microsoft initially licensed engines and technology from Dragon Systems. The real wonder is why Dragon Dictate allowed them to make other Dragons? If Dragon Dictate technology was "best", then IBM and Microsoft were not behind. While each is special and unique all Dragons come from common origin.

    Dragon Dictate blazed ahead to establish its' own new personal and professional markets. While IBM and Microsoft moved to accommodate their already established home and personal computing markets. Each using the same technology to develop their own unique products and prices for their respective markets. If Dragon Dictate Products are "good", then so too are IBM and Microsoft. From common engines and technical origins, the difference and function are more market -and- price. Not all Dragons are Dragon Dictate's.

    Over time the clear differentiation of priorities and territories, between market share and price targetting, have blurred and trended away. While the largest companies may "step on" the smallest, corporate conflict has been contained to technology and competition. All companies know that: "When Dragons Fight, Only The Earth Burns".

    DRAGONS SPEAK:
    Discrete, Continuous Or Natural

    Here you find ... Which Dragon Works For You?

  • Today your choice is Discrete or Continuous.
    Natural Speech has just begun to arrive.

  • If you have, and/or can afford, the computer "power" and software. You can afford Continuous Natural Speech.

  • If you do not have, and can not afford, both computer and software. You can afford Discrete and some Continuous Speech now.

  • If you can think and dictate over 100 -to- 150 words -per- minute. You need Continuous Natural Speech.

  • If you think and dictate 30 -to- 90 words -per- minute. You can use Discrete and Continuous Speech now.

  • If you talk without pause between word, phrase, and breath. You need Continuous Natural Speech.

  • If you can talk pausing between word, phrase, and breath. You can use Discrete and Continuous Natural Speech now.

  • If you talk slowly. You can use either Discrete and Continuous Speech now.

  • If you talk poorly. You may be able to use Continuous Natural Speech; or you may have to wait for Speech Technology to arrive.

  • If you depend upon your computer to input text and/or data. You can use Discret and Continuous Natural Speech now. You need to personally experience and compare, availability of preset commands, micros, hot keys, for each application. Both Dragon Dictate and IBM ViaVoice Gold are recommended.

  • If you need to dictation and operate your computer without Keyboard and Mouse. You need to personally experience and compare, availability of preset commands, micros, hot keys, for each application operating without a Keyboard and Mouse. Both Dragon Dictate and IBM ViaVoice Gold are recommended.

  • If you depend upon your computer to navigate, communicate, and input commands to applications. You can use Discrete and Continuous Natural Speech now. You need to personally experience and compare, availability of preset commands, micros, hot keys, for each application. Both Dragon Dictate and IBM ViaVoice Gold are recommended.

  • If you are presently running MS Dos and MS Windows 3.+ configuration: On a [now low powered] early "Pre-Pentium" and minimum memory. With a variety of Applications and/or Adaptive Technology. You may not be able to afford the change. Consider Discrete and Continuous Speech for Windows 3.+ Now, while the "Price Is Right" and the product is "Still Available".

  • If you are presently running MS Windows 95. On a [now low powered] early Pentium and minimum memory. With a variety of Applications and/or Adaptive Technology. You still may not be able to afford the change. Consider Continuous and Natural Speech For Windows 95. Now, while power reqirements are still low, or the "Price Is Right", and the product is "Still Available".

    BY THE WAY:

    DRAGONS ARE MYTHICAL,
    Unless They Live In Your Computer.

    TECHNO-MAGIC IS REAL.

    DRAGONS ARE ONLY DRAGONS,
    When Merlin Is The Technician.

    Top | ACSP Home | SuperAdaptoid Column

    06-06-97/01-06-98