IBM Voice Type Dictation was an early version of Speech Recognition Software from IBM.
In addition, IBM provided IBM MedSpeak Radiology, a reasonably successful program exclusively for Radiologists.
Their current Desktop Speech Recognition program, IBM Via Voice, is currently licensed to ScanSoft.

Following are a number of articles from 1996 through 2000 on various IBM Speech Recognition Software Programs.

IBM has had 3 updates for IBM Voicetype V 1.32


1.      UPDATE: IBM Voicetype 1.32A


        -Inludes several feature enhancements, as well as better navigation.



2.      UPDATE: IBM Voicetype 1.32A for 32-Bit PCMCIA card support for Windows 95.



        -If you are using Windows 95, with a PCMCIA card, and your card has 32-bit support.



3.      UPDATE: IBM Voicetype 1.32B



        -Some computers have had known problems with the voice-manager.

        -This corrects this problem.



You must install these updates.  Each, individually.  

You must have 3 Formatted Diskettes, and at least 2 MB of Hard Drive Space.



You must copy three files to a formatted high density 3 1/2" floppy on the A: Drive.

IBM HAS concentrated on two very important factors in building their speech recognition product. These are SPEED and ACCURACY.

Available in Windows 3.1, Windows 95, and OS/2 systems, IBM VoiceType also comes in several different languages, and it also has industry-specific vocabularies for Journalism, Legal and Medical entities.

The information contained here is for Voicetype 3.0 for Windows 95. Windows 3.x information is also available.

Add-Ons include Kolvox OfficeTALK & Kolvox LawTALK.

Features & Benefits

Now you can free yourself forever from using your computer in the conventional way and work smarter -- in a way that's more productive... simply by speaking. With VoiceType Dictation 3.0 for Windows® 95, you'll be able to dictate text directly into your applications and navigate your desktop or notebook computer... using just your voice.

Voice your independence
Speech is the most natural and effortless means of human communication, and thousands of people around the world are already computing with the power of speech, thanks to IBM's tried and tested VoiceType technology. Created for the Windows 95 environment, VoiceType Dictation 3.0 boasts second-to-none accuracy and speed, while setting a new standard for ease of use.

Because it's speaker-independent, VoiceType Dictation will recognize different speech patterns. Based on customer tests, it's also more than 90% accurate for most people right out of the box -- without enrollment. Plus, VoiceType Dictation will learn from you as you work with it. In fact, the more you use it, the better it gets! Although enrolling is optional, it can increase accuracy even further. More than one person can use it in a collaborative environment, by creating separate users.

Speaking of accuracy...
When we say that IBM VoiceType Dictation offers second-to-none accuracy, we can say it with confidence. That's because IBM's proven technology leads the industry. By using what we call 'isolated word' -- or speaking each word individually -- the program precisely translates your words into text. The VoiceType system not only searches for an acoustic match to any word you say, but can even differentiate between like-sounding words such as 'to,' 'two,' and 'too,' by analyzing them in context.

Let's talk productivity
When it comes to speed, VoiceType Dictation takes the lead. In fact, you can dictate between 70 and 100 words per minute -- faster than most people can type. You'll work even more quickly using VoiceType Dictation's 'macro' function -- a way of combining multiple steps with one command -- to create input shortcuts for frequently used blocks of text.

VoiceType Dictation lets you work with both your hands and eyes free, so you'll be able to interact with your applications easily, naturally and more productively.

The benefits are enormous: Because you don't have to keep your eyes on the screen, you can dictate out in the field or in your car. Research scientists can work with a microscope and dictate their findings simultaneously. A radiologist can dictate a report while reviewing X-rays. Attorneys can review cases and dictate pertinent information -- without ever taking their eyes from their papers.

Whatever your business, with VoiceType Dictation you'll enjoy the benefits that come from increased personal productivity. You'll also have the assurance that comes from knowing you can:

• Spell correctly -- without running spell check.
• Create letters, memos and confidential reports yourself -- even after hours -- with fast turnaround.
• Work anytime and anywhere you like -- at the kitchen table, in your hotel room, or out in the field -- with VoiceType Dictation on your notebook computer.
• Dictate text while you sit back, relax and rest your eyes.
• Make the most of your computer, even if you aren't an accomplished typist or can't use a keyboard.

Speak your mind
It's only natural that we've made IBM VoiceType Dictation 3.0 so easy to use. For example, the program doesn't require that you correct errors right away, so you can concentrate on what you're doing -- not on how to do it. You can make corrections whenever you like -- as you work or later on. This means you can use the system freely, with your eyes and hands free -- for example, to work with reference material. And since there is no need to watch your screen or break your train of thought to make corrections, you can sit back and relax and keep your mind on your dictation.

What's more, you can choose to save your work and have someone else make any necessary corrections later. Or play back an audio version of your dictation for editing and correction. It's that convenient.

VoiceType Dictation also lets you create your own templates for filling in forms quickly and easily with faster turnaround and fewer mistakes. With VoiceType Dictation and a notebook computer, architects, police and other professionals can quickly fill out forms in the field -- simply by speaking. Doctors' offices can enter patient information and progress notes. Others can use VoiceType Dictation as a productivity aid for sophisticated data-center operations.

You dictate the way you work
Now you can use VoiceType Dictation to create text in your favorite applications! You simply dictate directly into a program like Microsoft® Word. Using VoiceType Dictation with other applications couldn't be easier -- a quick dictation window pops up automatically to let you know if an application is not speech-aware.

The program gives you a wide variety of ways to dictate text and integrate your applications, including VoiceType Dictation's own speech-aware word processor, called VoicePad. You can also create your own dictation shortcuts -- called macros -- for often-used text or your return address, for example.

VoiceType Dictation comes with a large 22,000-word base vocabulary, to which you can add another 42,000 words of your own! If you like, you can choose to add your client name-and-address database, product names or other business-specific data to your personal vocabulary.

Talk is cheap!
You'll be on speaking terms with your computer right away because there's no proprietary hardware to install -- VoiceType Dictation 3.0 works with the standard Sound Blaster^TM-compatible or Mwave® sound card you may already have. Plus, the software is quick and easy to load from CD (or diskettes, if you prefer), so you can give your PC a good talking to in no time.

The price is extremely attractive, too... You get everything you need in one package -- including a high-quality microphone! It's a comfortable active-noise cancelling headset microphone that gives you freedom of movement while cutting out ambient noise. The earpiece speaker offers private audio feedback.

At your command
VoiceType Dictation's VoiceCenter is the hub from which you can easily control your computer with voice commands -- move around your desktop...start or switch applications...save or print a file -- all by speaking in your normal voice. Take advantage of the multitasking abilities of Windows 95 -- just use your own voice to switch between and control your favorite applications.

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 slightly more distinct way of speaking that leaves a one-tenth of a second gap between each word.

As you work with the VoiceCenter, the talk area displays what the computer heard you say -- or a polite "pardon me?" if it didn't understand you. The "what can I say" window will help you out if you're not sure of an exact command. You can also create custom voice commands or speed up often-repeated actions by combining them under a single command. Of course, if you wish, you can use your keyboard and mouse, too.

We speak your language
IBM VoiceType Dictation 3.0 is available in six languages -- French, German, Italian, Spanish, U.S. English. and U.K. English, with Chinese and Japanese coming soon.

From the IBM speech family
IBM offers the most advanced selection of speech products in the world. Along with VoiceType Dictation 3.0 for Windows 95, there are other versions for OS/2® and Windows® 3.1, and Windows 95 on the 486 and lower-end Pentium platforms. If you work in a particular profession, you can make the most of VoiceType Dictation with the VoiceType Professional Series products, complete with additional vocabularies specific to your vocation. Specialty vocabularies are also available separately for legal, medical, and journalism.

Highlights

• Dictate text directly into your favorite applications -- such as Microsoft Word. Control your computer simply by telling it what to do
• Save time and money -- letters and reports can be ready when you need them -- even after hours...or while traveling
• Get your work done fast -- the program can take your dictation more efficiently and more quickly than most people can type, with 'second-to-none' accuracy
• Concentrate on what you're doing -- not how you're doing it. With VoiceType Dictation 3.0, your hands and your eyes are free for other tasks
• Make the most of the large 22,000-word base vocabulary, plus a customizable, 42,000-word personal vocabulary
• Work without breaking your train of thought or looking at the screen, thanks to delayed correction capability
• Play back what you've dictated for easy editing and proofing
• Declare speaker independence! Based on customer tests, VoiceType Dictation 3.0 is more than 90% accurate for most people right out of the box -- enrollment is not required
• Be on speaking terms with your computer right away -- the program runs with standard Sound Blaster -- compatible or Mwave sound card implementations.

System Requirements

 Minimum Pentium 90 MHz processor and a 256KB L2 cache or faster processor
 16 MB of RAM (recommended 20 MB)
 33 MB dedicated hard disk space | additional 30 MB when enrolling and additional 30 MB for specialty vocabularies.
 Microsoft Windows 95

 

---

 

Unleash the Power!
IBM Voice Type Dictation 3.1

IBM® has concentrated on two very important factors in building their speech recognition product. This is SPEED and ACCURACY.

Available in Windows 3.1, Windows '95 and OS/2 systems, IBM Voicetype also comes in several different languages, and it also has industry specific vocabularies for Journalism, Legal and Medical entities.

Add-Ons include Kolvox OfficeTALK & Kolvox LawTALK.

Voice-to-text capability

• Facilitates the creation of documents
• Saves time and increases productivity by eliminating transcription
• Provides a solution for people who are unable (or unwilling) to use a keyboard

Hands-free, eyes-free operation

• Enables users in technical, production and office environments to dictate text, product reports or fill in forms while using their hands and eyes for other tasks
• Physicians and other medical personnel can read x-rays or examine patients and describe findings simultaneously

Large General-language vocabulary

• Provides a ready-to-use language model designed to serve a wide range of professional and business users
• Helps ensure accurate word capture
• Allows you to customize the system easily by adding up to 2,000 words to the vocabulary

Optional industry-specific vocabularies

• Ready-to-use solutions for radiology, emergency medicine, journalism and legal enhance applications
• Specialized terminology and word-pattern models enable the system to frequently recognize words that are run together, slurred or partially obscured by background noise
• Support for additional industry-specific vocabularies is continually being expanded

Speaker enrollment function

• Maximizes dictation accuracy by creating a model for every individual speaker

Clipboard data transfer

• Transfer dictated text to OS/2 and Windows applications by cut-and-paste, using voice, keyboard or mouse commands
• Use keyboard emulation to transfer data into applications that do not support cut-and-paste

Navigation controls

• Use voice commands to operate OS/2, Windows and compatible applications
• Includes automatic "speech enabling" of the OS/2 WorkPlace Shell and Presentation Manager controls, as well as Windows icons and tool bars

Voice Action Editor

• Allows users and developers to define macros that perform a sequence of actions with a single voice command
• Facilitates development of customized, application-specific solutions
• Increases productivity by letting users insert prerecorded phrases or paragraphs of repetitive text with a single spoken word

PCMCIA VoiceType Dictation Adapter

• Supports notebook computers for transportable dictation solutions
• Helps make it easy for people to use the system, wherever they need to work

Multiple language support

• Available for OS/2 and Windows in U.S. English, U.K. English, French, German, Italian and Spanish.

---

 

 

Unleash the Power with the Medical Vocabularies!
 

IBM® has concentrated on two very important factors in building their speech recognition product. This is SPEED and ACCURACY. Not as navigational as the other products, but then, if you don't need navigation and an extremely large vocabulary, this i s a tough product to beat.

The Radiology Edition contains a 19,000-word vocabulary of medical terminology for radiology and related specialties. Derived by analyzing reports from several major hospitals in the United States, it is specifically construc ted for use by radiologists while reading x-rays or films and contains words and medication names commonly used in a radiology setting.

The Emergency Medicine Edition is a 16,000-word vocabulary of medical terms commonly used in a hospital emergency room environment. It can also be used effectively by other doctors and health-care professionals.

Available in Windows 3.1, Windows '95 and OS/2 systems, IBM Voicetype also comes in several different languages, and it also has industry specific vocabularies for Journalism, Legal and Medical entities. Add-Ons include Kolvox OfficeTALK & Kolvox LawTALK.

Voice-to-text capability

• Facilitates the creation of documents
• Saves time and increases productivity by eliminating transcription
• Provides a solution for people who are unable (or unwilling) to use a keyboard

Hands-free, eyes-free operation

• Enables users in technical, production and office environments to dictate text, product reports or fill in forms while using their hands and eyes for other tasks
• Physicians and other medical personnel can read x-rays or examine patients and describe findings simultaneously

Large General-language vocabulary

• Provides a ready-to-use language model designed to serve a wide range of professional and business users
• Helps ensure accurate word capture
• Allows you to customize the system easily by adding up to 2,000 words to the vocabulary

Optional industry-specific vocabularies

• Ready-to-use solutions for radiology, emergency medicine, journalism and legal enhance applications
• Specialized terminology and word-pattern models enable the system to frequently recognize words that are run together, slurred or partially obscured by background noise
• Support for additional industry-specific vocabularies is continually being expanded

Speaker enrollment function

• Maximizes dictation accuracy by creating a model for every individual speaker

Clipboard data transfer

• Transfer dictated text to OS/2 and Windows applications by cut-and-paste, using voice, keyboard or mouse commands
• Use keyboard emulation to transfer data into applications that do not support cut-and-paste

Navigation controls

• Use voice commands to operate OS/2, Windows and compatible applications
• Includes automatic "speech enabling" of the OS/2 WorkPlace Shell and Presentation Manager controls, as well as Windows icons and tool bars

Voice Action Editor

• Allows users and developers to define macros that perform a sequence of actions with a single voice command
• Facilitates development of customized, application-specific solutions
• Increases productivity by letting users insert prerecorded phrases or paragraphs of repetitive text with a single spoken word

PCMCIA VoiceType Dictation Adapter

• Supports notebook computers for transportable dictation solutions
• Helps make it easy for people to use the system, wherever they need to work

Multiple language support

• Available for OS/2 and Windows in U.S. English, U.K. English, French, German, Italian and Spanish.

 

 

---

 

Unleash the Power with the Legal Vocabulary!
 

IBM® has concentrated on two very important factors in building their speech recognition product. This is SPEED and ACCURACY. Not as navigational as the other products, but then, if you don't need navigation and an extremely large vocabulary, this i s a tough product to beat.

VoiceType Vocabulary for Legal Dictation

VoiceType Vocabulary for Legal Dictation consists of 25,000 words specific to the legal profession and likely to be used in a law office. It supports most legal abbreviations and popular forms of legal citation. The vocabulary was developed by a stat istical analysis of numerous court cases and legal reference materials, including some provided by West Publishing.

Available in Windows 3.1, Windows '95 and OS/2 systems, IBM Voicetype also comes in several different languages, and it also has industry specific vocabularies for Journalism, Legal and Medical entities. Add-Ons include Kolvox OfficeTALK & Kolvox LawTALK.

Voice-to-text capability

• Facilitates the creation of documents
• Saves time and increases productivity by eliminating transcription
• Provides a solution for people who are unable (or unwilling) to use a keyboard

Hands-free, eyes-free operation

• Enables users in technical, production and office environments to dictate text, product reports or fill in forms while using their hands and eyes for other tasks
• Physicians and other medical personnel can read x-rays or examine patients and describe findings simultaneously

Large General-language vocabulary

• Provides a ready-to-use language model designed to serve a wide range of professional and business users
• Helps ensure accurate word capture
• Allows you to customize the system easily by adding up to 2,000 words to the vocabulary

Optional industry-specific vocabularies

• Ready-to-use solutions for radiology, emergency medicine, journalism and legal enhance applications
• Specialized terminology and word-pattern models enable the system to frequently recognize words that are run together, slurred or partially obscured by background noise
• Support for additional industry-specific vocabularies is continually being expanded

Speaker enrollment function

• Maximizes dictation accuracy by creating a model for every individual speaker

Clipboard data transfer

• Transfer dictated text to OS/2 and Windows applications by cut-and-paste, using voice, keyboard or mouse commands
• Use keyboard emulation to transfer data into applications that do not support cut-and-paste

Navigation controls

• Use voice commands to operate OS/2, Windows and compatible applications
• Includes automatic "speech enabling" of the OS/2 WorkPlace Shell and Presentation Manager controls, as well as Windows icons and tool bars

Voice Action Editor

• Allows users and developers to define macros that perform a sequence of actions with a single voice command
• Facilitates development of customized, application-specific solutions
• Increases productivity by letting users insert prerecorded phrases or paragraphs of repetitive text with a single spoken word

PCMCIA VoiceType Dictation Adapter

• Supports notebook computers for transportable dictation solutions
• Helps make it easy for people to use the system, wherever they need to work

Multiple language support

• Available for OS/2 and Windows in U.S. English, U.K. English, French, German, Italian and Spanish.

 

---

Unleash the Power with the Journalism Vocabulary!

IBM® has concentrated on two very important factors in building their speech recognition product. This is SPEED and ACCURACY. Not as navigational as the other products, but then, if you don't need navigation and an extremely large vocabulary, this i s a tough product to beat.

VoiceType Vocabulary for Journalism Dictation

The Journalism add-on library is a 30,000-word vocabulary constructed from the archives of major newspapers. It contains a variety of significant proper names as well as other useful editorial terminology.

Available in Windows 3.1, Windows '95 and OS/2 systems, IBM Voicetype also comes in several different languages, and it also has industry specific vocabularies for Journalism, Legal and Medical entities. Add-Ons include Kolvox OfficeTALK & Kolvox LawTALK.

 

Voice-to-text capability

• Facilitates the creation of documents
• Saves time and increases productivity by eliminating transcription
• Provides a solution for people who are unable (or unwilling) to use a keyboard

Hands-free, eyes-free operation

• Enables users in technical, production and office environments to dictate text, product reports or fill in forms while using their hands and eyes for other tasks
• Physicians and other medical personnel can read x-rays or examine patients and describe findings simultaneously

Large General-language vocabulary

• Provides a ready-to-use language model designed to serve a wide range of professional and business users
• Helps ensure accurate word capture
• Allows you to customize the system easily by adding up to 2,000 words to the vocabulary

Optional industry-specific vocabularies

• Ready-to-use solutions for radiology, emergency medicine, journalism and legal enhance applications
• Specialized terminology and word-pattern models enable the system to frequently recognize words that are run together, slurred or partially obscured by background noise
• Support for additional industry-specific vocabularies is continually being expanded

Speaker enrollment function

• Maximizes dictation accuracy by creating a model for every individual speaker

Clipboard data transfer

• Transfer dictated text to OS/2 and Windows applications by cut-and-paste, using voice, keyboard or mouse commands
• Use keyboard emulation to transfer data into applications that do not support cut-and-paste

Navigation controls

• Use voice commands to operate OS/2, Windows and compatible applications
• Includes automatic "speech enabling" of the OS/2 WorkPlace Shell and Presentation Manager controls, as well as Windows icons and tool bars

Voice Action Editor

• Allows users and developers to define macros that perform a sequence of actions with a single voice command
• Facilitates development of customized, application-specific solutions
• Increases productivity by letting users insert prerecorded phrases or paragraphs of repetitive text with a single spoken word

PCMCIA VoiceType Dictation Adapter

• Supports notebook computers for transportable dictation solutions
• Helps make it easy for people to use the system, wherever they need to work

Multiple language support

• Available for OS/2 and Windows in U.S. English, U.K. English, French, German, Italian and Spanish.

 

 Minimum 486/SX/SLC 25 MHz (recommended 486DX2/66)
 16 MB of RAM (recommended 24 MB)
 33 MB dedicated hard disk space | additional 30 MB when enrolling
 OS/2 Ver. 2.1 or higher for OS/2 configuration
 Microsoft Windows 3.1x or Windows 95

---

IBM Via Voice for the MacIntosh

Easy -- In just minutes, ViaVoice can learn to recognize your voice, then it's off to the races. The intuitive interface, easy online help and an easy-to-understand User's Guide ensure you'll get the most out of ViaVoice -- and your Mac -- every day.

Natural -- the Mac is really intuitive and what's more intuitive than speaking? So why tie yourself to a keyboard? ViaVoice can work with you, learning as you use it, tailoring itself to your unique speech patterns to enhance voice recognition accuracy.

Fun -- OK, maybe not everything you do on your computer is for fun. But with ViaVoice for Mac, even such tasks as an overdue report, or the letter you owe Aunt Edna, can become faster, easier and a lot more interesting. And because more than one person can use ViaVoice for Mac on the same computer, maybe you can get someone else to dictate that letter . . . Whoa, this really is fun!

And we're only getting started . . .

	Setup Assistant can create a Personal Voice Model for each system user, ensuring the greatest voice recognition accuracy.
	"Analyze My Voice" teaches ViaVoice your speech patterns.
	"Analyze My Documents" teaches ViaVoice the context in which you use certain words and phrases.
	Dictation Macros create "shortcuts" for recognizing your most commonly used expressions.
	Automatically learns how you speak and write, improving accuracy as you use it
	Build personal vocabulary by Analyzing your Documents
	Use Apple PlainTalk® technology to read the text of your documents aloud
	Dictation into IBM SpeakPad
	Modeless operation when dictating (say voice commands and dictation interchangeably, without stopping dictation or requiring an attention word)
	Natural dictation of numbers, dates, times, measures, and prices
	Dictate letters, reports, and e-mail and then transfer them to your favorite Mac program
	Multiple ways to correct and edit text
	Deferred and delegated correction
	View your personal vocabulary and delete unneeded words using 'Manage My Vocabulary'

ViaVoice for Mac provides users with an intuitive, easy-to-use interface to perform everyday computer tasks, such as creating e-mail and editing documents, by speaking. The software is available this week on store shelves, online and in catalogs in time for the holiday buying season.

"IBM listened to the thousands of requests from Mac enthusiasts and brought its leading ViaVoice speech recognition technology to one of the most popular operating platforms," said Krishna Nathan, director of consumer voice systems, IBM Speech Systems. "ViaVoice for Mac is part of our continuing effort to make it easier and more natural for people to interact with their computer. IBM is the only company to offer its speech recognition technology on all of the most popular desktop operating platforms -- Windows, Linux and Macintosh."

"We are thrilled IBM is bringing its award-winning speech recognition software to the Mac," said Clent Richardson, Apple's vice president of Worldwide Developer Relations. "ViaVoice for Mac makes the most user-friendly Mac even easier to use."

The ViaVoice Advantage
Based on IBM's award-winning speech recognition technology, ViaVoice for Mac speech recognition software was created with simplicity in mind. The software incorporates an easy-to-use interface specifically designed for the Mac platform -- making it both easy and natural for all Mac users to conduct common computing tasks. ViaVoice for Mac provides users with the ability to create, edit and format documents in the most natural way. After text is created, the software's integrated transfer feature lets users transfer text into popular Macintosh applications, such as AppleWorks or Microsoft Word. Users can create voice shortcuts for commonly-used phrases by literally "calling" them out with the software's dictation macro editor. ViaVoice for Mac includes a 64,000 active word vocabulary, the largest on the market for Macintosh, and also integrates Apple's text-to-speech function that lets Apple computers read text back to the user. Also included with the product is a newly designed noise cancelling microphone for Mac users. This new mic has changeable colored accents to coordinate with all the popular iMac flavors.

Power for ViaVoice for Macintosh
IBM's ViaVoice for Mac runs on Mac OS 8.5.1, 8.6 and 9.0. The software requires a 233 MHz Power PC processor or higher, 48 MB RAM, 200 MB of hard disk space and a CD-ROM drive. The software runs on iMac, PowerMac G3 and G4, and PowerBook G3 computers, and requires an audio input jack that is compatible with Andrea NC-71 microphones.


About IBM Speech Systems
IBM speech recognition technology is based on 30 years of speech recognition research and development. With a portfolio of more than 100 patents, an established R&D business that includes researchers worldwide and a robust, proven speech technology available in many languages, IBM is a leader in speech recognition.


EDITORS NOTE:

The price noted herein is IBM's suggested retail price only which is subject to change without notice. Dealer prices may vary.

IBM, Power PC and ViaVoice are trademarks or registered trademarks of International Business Machines Corporation in the US. and /or other countries. Apple, AppleWorks, PowerBook, Mac, and Macintosh are trademarks of Apple Computer, Inc. registered in the US. and other countries. iMac and Power Mac are trademarks of Apple Computer, Inc.

Windows is a registered trademark of Microsoft Corporation in the United States, other countries or both.

Other company, product, and service names may be trademarks or service marks of others.

---