Speech Solutions
Text-to-Speech and Optical Character Recognition (OCR) Systems
Text-to-speech software is used to convert words from a computer document (e.g. word processor document, web page) into audible speech transmitted through the computer's speakers.
Text-to-speech technology can be integrated with Optical Character Recognition systems (OCR). OCRs can either be a self-contained unit (stand-alone) or part of a PC-based system. A stand-alone OCR or the scanner of a computer-based OCR system looks something like a small photocopier. It has a glass screen on which print to be "read" is placed. The scanner camera scans the material and, in a few seconds, the page is read out loud using synthetic voice. OCRs can read almost any book, newspaper or other typewritten materials, but they cannot read handwritten material.
When hooked up to a PC, scanned material can be converted to Braille, large print, voice, computer accessible files, or other media.
Ease of Use
Even people who have never used a computer can learn to use a stand-alone OCR system with only a few hours of training. People who already know how to use a standard computer require modest training to learn to use a computer-based OCR system.
Pricing Information
An OCR self-contained unit starts at about $4,000. OCR systems, which include a scanner and PC software, begin at about $1,100 (not including the cost of the PC).
Some Vendors and Products
Windows-based Text-to-Speech software:
- Next Up Technology: TextAloud MP3
Stand Alone Text-to-Speech Tools:
- Wizcom Technologies, Inc: Quick Link Pen
Optical Character Recognition Systems:
- Freedom Scientific: OpenBook
- Kurzweil Educational Systems: Kurzweil 1000
- Pulse Data HumanWare, Inc: ScannaR
- Robotron Group: Galileo Reading System, Pronto Reading Machine, Rainbow Reading Machine
- TeleSensory Corporation: Reading Edge
Screen Readers
A Screen Reader is the commonly used term for Voice Output Technology, in which hardware and software combine to produce synthesized voice output for text displayed on the computer screen, as well as for keystrokes entered on a keyboard. In a DOS environment, synthetic voice must be produced by a hardware synthesizer specially designed for that purpose. In a Windows environment, you can use software synthesizer programs, which make use of the sound system currently included with most PCs. These are less expensive than hardware synthesizers. However, in all environments, screen-reading software is needed to direct the hardware/software speech synthesizer. Before you purchase, be sure that the synthesizer and screen reading software are compatible.
Ease of Use
Training in keystroke commands and specific programs is required to use these programs successfully. To aid with training, companies have produced tutorials in braille, large print, audiocassette, and CD-ROM formats. The subjects of these tutorials include: various versions of the Windows operating system, MS Word (97 and 2000), Outlook 2000, Excel, WordPerfect for Windows, Netscape, and Internet Explorer version 5.
Pricing Information
The price of synthetic voice usually depends on the quality of the speech generated and whether the synthesizer is a piece of hardware or a software application. Hardware synthesizers cost between $300 and $1200. Some software synthesizer applications are provided with many screen readers at no additional cost. Other software synthesizers are priced at approximately $200 (These prices do not include the cost of the PC).
Prices for screen reading software vary. Screen readers for Windows-based systems cost between $600 and $1,400. Screen-reading software for DOS systems range in price from free to $600. Prices for tutorials range from $60 to $85.
Some Vendors and Products
- ALVA Access Group, Inc: OutSpoken for Windows or Macintosh
- Artic Technologies: WinVision
- Dolphin Computer Access: Hal, Supernova
- Freedom Scientific: JAWS
- GW Micro: Window-Eyes
Talking Internet Browsers
Although the majority of screen readers work well with mainstream Internet browsers, there are software packages currently on the market that use the same technology as screen reading software, but are designed specifically for surfing the Internet. The packages include a software speech synthesizer, an Internet browser, and an easy to use set of navigation and reading keystrokes. This software is primarily designed for people who do not wish to learn Windows navigation, nor other applications, but who want to surf the web.
Ease of Use
Some training is required in order to use these packages.
Pricing Information
These software packages range in price from $150 to $250.
Some Vendors and Products
- IBM: Home Page Reader
- Serotek Corporation: FreedomBox
Voice Recognition
Voice Recognition allows a user to use his or her voice as an input device. Voice recognition may be used to dictate text into the computer or to give commands to the computer (such as opening application programs, pulling down menus, or saving work).
Older voice recognition applications require each word to be separated by a distinct space. This allows the machine to determine where one word begins and the next stops. This style of dictation is called discrete speech. Many people (especially those with learning disabilities) prefer these systems to the newer continuous speech.
Continuous speech voice recognition applications allow a user to dictate text fluently into the computer. These new applications can recognize speech at up to 160 words per minute. While these systems do give the user system control they are not yet hands free.
Voice recognition uses a neural net to "learn" to recognize your voice. As you speak, the voice recognition software remembers the way you say each word. This customization allows voice recognition, even though everyone speaks with varying accents and inflection.
In addition to learning how you pronounce words voice recognition also uses grammatical context and frequency of use to predict the word you wish to input. These powerful statistical tools allow the software to cut down the massive language database before you even speak the next word.
While the accuracy of voice recognition has improved over the past few years some users still experience problems with accuracy either because of the way they speak or the nature of their voice.
Ease of Use
These packages are fairly easy to install and use. They require relatively little training for proper use.
Pricing information
The prices for these applications range from $600 to $1000.
Some Vendors and Products
Voice recognition for Windows:
- ScanSoft, Inc: Dragon NaturallySpeaking, ViaVoice
Voice Recognition for Macintosh:
- ScanSoft, Inc: ViaVoice for Mac