Page 466 - Biomedical Engineering and Design Handbook Volume 2, Applications
P. 466
444 REHABILITATION ENGINEERING AND PROSTHETICS DESIGN
capacitive switches used on some elevator buttons. Bioelectrical switches detect muscle electrical
activity (EMGs) or eye movement (EOG) by attaching electrodes to the skin.
Another type of body-generated signal is respiration that is detected by measuring either airflow
or air pressure using what is called a sip-and-puff switch activated by blowing air into the switch or
sucking air out of it. Arrays of single puff-and-sip switches are often used for multiple functions. The
most common use has been two puff-and-sip switches mounted side by side for wheelchair control.
Puffing on both cause forward motion, sipping on both leads to reverse, and puffing on only one
results in a turn in that direction.
15.2.2 Control Interfaces for Direct Selection
The most common control interface for direct selection is the keyboard. However, there are many
different types of keyboards and each requires unique user skills. Some consumers have limited
range of movement, but good fine motor control. In this case, a contracted keyboard (closely spaced
keys over a small range) can be more effective than the standard keyboard. Other individuals have
difficulty accessing small targets, but they have good range of movement. In this case, an expanded
keyboard, in which the size of each key can be up to several inches, may allow direct selection.
Different keyboard layouts can also be used (e.g., for left-hand- or right-hand-only typing). It is also
possible to make the keys in different sizes and different shapes on the same keyboard. Touch screens
allow a user to choose from the selection set by pointing directly to the item on the screen. Because
the item chosen is the one that is pointed at, this method can be easier for some users to understand.
Automatic speech recognition (ASR), in which the individual uses sounds, letters, or words as a
selection method, is another alternative to keyboard input. In most systems the speech recognition is
speaker-dependent, and the user trains the system to recognize his voice by producing several
samples of the same element (Comerford et al., 1997). ASR system use is increasing in the main-
stream commercial market on the Internet, for dictation, for cell phone use, for personal digital assi-
∗
tants (PDAs), and for most other computer activities. Microsoft Vista includes ASR as part of the
built-in package of accessories. Persons with disabilities will be the beneficiaries of this expanded
usage of ASR systems. These systems all use continuous ASR techniques in which the user can
speak in almost normal patterns with sight pauses between words.
Speaker-independent systems recognize speech patterns of different individuals without training
(Gallant, 1989). These systems are developed using samples of speech from hundreds of people and
information provided by phonologists on the various pronunciations of words (Baker, 1981). The
total recognition vocabulary is generally small. Discrete ASR systems are used in assistive technol-
ogy applications for wheelchair control and electronic aids to daily living for appliance control.
These systems require the user to pause between words and result in very abnormal speech patterns.
They are only used in limited applications requiring a few commands to be recognized.
Often the microphones supplied with ASR systems are not adequate when the user has limited
breath support, special positioning requirements, or low-volume speech (Anson, 1997). Many indi-
viduals who have disabilities may not be able to independently don and doff headset microphones
that are normally supplied with commercial ASR systems. In these cases, desk-mounted types are
often used. Current ASR systems utilize commonly available sound cards rather than separate hard-
ware installed in the computer (Anson, 1999).
15.2.3 Control Interfaces for Indirect Access
Indirect methods of selection use a single switch or an array of switches. Cook and Polgar (2008)
describe a variety of interfaces that are used for indirect selection.
∗
Microsoft, Seattle, Wash.

