Options to Consider: Speech Recognition
JAN’s Technical Series is designed to help employers determine effective accommodations and comply with Title I of the Americans with Disabilities Act (ADA). Each publication in the series addresses a specific medical condition and provides information about the condition, ADA information, accommodation ideas, and resources for additional information.
The Technical Series is a starting point in the accommodation process and may not address every situation. Accommodations should be made on a case by case basis, considering each employee’s individual limitations and accommodation needs. Employers are encouraged to contact JAN to discuss specific situations in more detail.
For information on assistive technology and other accommodation ideas, visit JAN's Searchable Online Accommodation Resource (SOAR) at http:/askjan.org/soar.
Individuals with physical, cognitive, sensory, and learning impairments may have difficulty accessing computers. Traditionally, a computer user accesses a computer with a standard point and click mouse and a QWERTY keyboard (named for the top left-hand side of the rows). The computer takes the information that is inputted and processes it. Individuals with physical, sensory, or developmental limitations may not be able to use these standard input devices effectively and may benefit from using speech recognition software. Standard computer input devices are keyboards made for typists who use two hands and ten fingers; speech recognition software is made for computer users with a variety of limitations, including individuals with no hand or finger movement.
Speech recognition technology has several components: noise-canceling input, a speech recognition engine, vocabularies, application interfaces, and rudimentary natural-language processing. In contrast, voice recognition refers to voice-print security systems, commonly called voice ID. These terms are often used interchangeably, although they have different meanings. This document addresses speech recognition options. There are two classes of speech recognition technology: speaker-dependent, where the user has to train the system to recognize the user's voice, and speaker independent, where training the system is not required. There are two categories of speech recognition: keyboard and keypad. Keyboard applications allow users to speak directly to their computers, complementing or replacing the keyboard. Keypad applications use speech to replace the telephone keypad as input for accessing voice mail and navigating a telephone system's menus. They also allow the telephone to act as a remote computer peripheral. In the past, users may have needed to choose between discrete and continuous speech recognition engines. Discrete systems require users to pause between each word while continuous systems allow users to speak more "naturally." More sophisticated and modern software accepts natural speech for applications such as call routing, speech-to-text, voice dialing, and voice search. Technological advances in the area of speech recognition have made software and devices more user-friendly, available, and accurate.
This document is a non-inclusive list of speech recognition systems. If you would like additional information on devices such as ergonomic, miniature, expanded, and one-handed keyboards; touchpads; trackballs; joysticks; switches; handwritten entry; scanners; alternative mice; augmentative communication devices; or alternative input devices and software, contact the Job Accommodation Network. For a list of vendors, visit: http://soar.askjan.org/solution/254
Windows speech recognition is a built-in accessibility feature available on newer versions of Windows (XP, 7 & 8). Speech recognition allows the user to operate the PC and dictate or edit text in most applications. Windows offers a tutorial, command suggestions, and a speech dictionary for adding words, topping a specific word from being dictated, or correcting or deleting a word in the dictionary. A microphone is required in order to use Windows speech recognition.
Dragon NaturallySpeaking is speech recognition software that enables users to control their computer, fill out forms, create and edit documents, reports, spreadsheets, and e-mail-all by voice. Dragon Naturally Speaking is integrated with Microsoft Word, Open Office Writer, Microsoft Excel, Microsoft Outlook, Microsoft PowerPoint, Microsoft Internet Explorer, Mozilla Firefox, Google Chrome and Corel WordPerfect. It works with virtually all Windows-based applications and Gmail and Hotmail/Outlook.com email applications. Documents can be reviewed with the text-to-speech option available and users can customize vocabulary, characters, and commands to suit their specific needs. When away from your PC, speech can be captured by a Nuance-approved digital recorder and Dragon can transcribe the audio file when you return to your PC.
Dragon Dictate is speech recognition software developed for Macintosh OS that allows users to create and edit documents, manage email, and access the Internet and social networks by voice. Users can create custom voice commands or capture ideas using a digital voice recorder that can be transcribed at a later time.
MacSpeech's Scribe transcribes speech that is saved as an spoken-word audio file to text for Macintosh OS X. MacSpeech Scribe supports a variety of audio file formats, including .wav, .aif, .aiff, .m4v, .mp4, and .m4a and has the ability to create up to six individual speech profiles.
MacSpeech Dictate Medical and Mac Speech Dictate Legal offer support to those in healthcare or legal occupations. MacSpeech Dictate Medical supports and fully understands over 54 medical and dental disciplines and specialties. MacSpeech Dictate Legal supports and fully understands over 30,000 legal words and terms.
1 Infinite Loop
Cupertino, CA 95014
VXI Corporation's CT Computer/Telephone Switch allows users to toggle back and forth between the phone and PC with a push of a button. It can also be used to listen to a call and computer prompts simultaneously and is compatible with most phones.
The Andrea Electronics Personal Computer Telephone Interface PCTI-3 Way allows headset users to connect simultaneously to their computers and their telephones for communication while computing. The device works with most telephones and an Andrea PC headset, which can be used instead of the phone handset. The user can switch between the PC and phone or use both.
Synapse TAP Workstations
Synapse TAP Workstations can offer a solution to compatibility issues encountered in network, mainframe, emulations or Unix environments. The Synapse TAP workstation (http://www.unixspeech.com/tap1host.htm) can be plugged into any host computer the workstation supports including Sun Microsystems, Silicon Graphics, HP, UNIX, Macintosh, PC operating systems and mainframe environments. The Synapse TAP device converts keystroke and mouse events originating in the speech-accessing computer and converts them into understandable events before presenting them to the host computer. Because the host computer is receiving only keystrokes and mouse events no issues of compatibility can develop.
Additional TAP devices are required for each host computer and the addition of a Synapse TAP Switch is necessary to attach a Sun, SGI, Mac, or PC Host. For more information on the Synapse TAP workstation visit: http://www.unixspeech.com/Default.htm
MediDox uses structured templates and speech recognition and/or handwriting recognition to enter patient information, progress notes, referral letters, prescriptions, and reports.
J-Say Pro brings together Dragon NaturallySpeaking Professional, from Nuance, and JAWS For Windows, from Freedom Scientific. With the integration of these two applications, the user is able to dictate text, use their voice as a means of computer input and control, or use their Braille display while having the software scan and read text back. Complete access to this application is made possible enabling people, without having to touch the keyboard or look at the screen, to create documents, control text layout, fill in forms and other complex documents, and open/save files, amongst other things. J-Say way takes away the need to use either one's hands or one's eyes; users can do it all by controlling the application with their voice and accessing the computer screen with their ears.
J-Say Pro is directly compatible with Windows XP Home & Pro, Windows Media Centre Edition, Windows Vista & 7, Microsoft Office 2007 or 2010, JAWS for Windows 14, and Dragon Naturally Speaking Professional 12.