We engaged in a collaborative effort with TeleSensory Systems, Inc., to develop a new TeleBraille -- a TDD for the Deaf-Blind using Braille output. (The old TeleBraille was no longer produced due to the discontinuance of the TDD upon which it was based.) The goal was to develop a state-of-the-art system, with added modern features demanded by deaf-blind consumers, and with a view to easier transfer to other TDD units should the one selected ever be discontinued. All development work was conducted at Smith-Kettlewell by Tom Fowle of our RERC staff, and production engineering and final manufacture will be undertaken by TeleSensory Systems.
Extensive input was sought from deaf-blind TeleBraille users regarding the features they consider desirable in a new telecommunications device, and preliminary design specifications were arrived at. The Ultratec Supercom proved suitable as a TDD to base the design on, as it provided all TDD functions in a single package. Ultratec was most cooperative in making necessary software and hardware modifications to allow access to all information from the Supercom display, and in working with our staff to develop a standard for interfacing the Supercom to the TeleSensory Navigator hardware (which provided the braille display output for the system). After completion and intensive in-house testing of the prototype, the device was successfully transferred to TeleSensory Inc for commercial production. Mr. Fowle produced an updated user manual for the system.
An abbreviated list of some of the new features follows.
We gratefully acknowledge the following agencies for their time, effort, skill and diligence in administering the Smith-Kettlewell Deaf-Blind Technology Needs Questionnaire to their clients: Center for Deaf-Blind Persons, Milwaukee; FIND (Functional Independence Training, Inc.), Minneapolis; Hawaii Vocational Rehabilitation and Services for the Blind, Dept. of Social Services, Honolulu; Interpreting School of Seattle; Massachusetts Commission for the Blind, Boston; Virginia Department for the Visually Handicapped, Richmond; and the Arkansas Department of Human Services, Division of Services for the Blind.]
We completed a survey of deaf-blind individuals to determine their opinions on technological needs. Thanks to the cooperation of a large number of agencies throughout the United States* we received 61 completed Deaf-Blind Technology Needs Questionnaires. The following is a brief summary of the survey results.
This project addresses the need for a mechanical hand capable of performing the fingerspelling code used for communication with persons who are deaf and blind. We have collaborated with Upstart Robots of San Francisco in pursuing the further development and technology transfer of this project. Upstart obtained a SBIR grant, with Smith-Kettlewell as consultant, to develop a Dexter III prototype with a view toward ultimate commercial production. Dexter III can form all 26 letters of the alphabet in a way which appears to the researchers' visual observations to be good simulations of the human hand shapes, or close enough to not be confused with any other letters. Wrist flexion has been incorporated to allow formation of letters G, H, and P, as well as formation of the "dynamic" letters, J and Z. Letters transition smoothly, without needing to move to a neutral position between letter pairs.
Five deaf-blind tactile fingerspelling readers helped test variations of Dexter III before the end of 1993. Among other tests, ten meaningful sentences were presented to each evaluator. The evaluator's responses, which were either in the form of fingerspelling or signing, were recorded both on paper and on videotape. A confusion matrix was synthesized to plot errors and help determine which letters needed improvement in their clarity of formation. Results were encouraging and led to a number of suggestions for further improvement.
The simple need to know when to take medication sometimes wreaks havoc within a deaf-blind person's family, as it requires an able-bodied person to be in regular physical proximity to the patient. Part of our Deaf-Blind Technology Needs Questionnaire dealt with the deaf-blind population's perceived need for a medication reminder which could be used without vision or hearing. Of the 61 respondents, 41 indicated they take medication. Eighteen (nearly half) of those individuals indicated they would like a vibrating reminder. A vibrating reminder responsive to these needs could also be utilized for any other scheduling/reminder functions in the daily life of the deaf-blind consumer. A
ccordingly, we surveyed commercially available devices which might be usable by deaf-blind persons. We obtained assistance from Ms. Paula Guerette of the Rehabilitation Research and Training Center on Aging at Rancho Los Amigos Medical Center in Downey, California, who has done an extensive survey of medication dispensers and reminders, and shared the human factors details of these with us. Not one device identified could be used independently by a deaf-blind person.
We therefore initiated, in collaboration with Mr. David Herrmann, a graduate of the San Francisco State University Department of Design and Industry, exploration of possible designs for a vibrating medication reminder which could be used independently by deaf-blind people. As a result, Mr. Herrmann completed the fabrication of a functioning prototype which served as a class project. The device incorporates two clock faces superimposed; one has tactile markings to indicate the numbers, and the two clock hands may be felt. The other clock face has a two-position switch at each number location. Moving a switch away from the center automatically sets the unit to vibrate at that time. (An AM-PM switch was not incorporated.) This prototype received very favorable comments from initial blind testers, who pointed out that hearing blind persons would also find a vibrotactile reminder valuable. Further evaluation and development work is planned.
Responding to a joint request from the Arizona State Department of Rehabilitation and an on-staff audiologist at the Arizona State School for the Deaf and Blind, a hearing-aid battery tester was developed for blind hearing aid wearers. A visual "Button Battery Tester" made by Activair (a division of Mallory) was adapted by adding a vibrotactile output. The prototype accommodates two tactile transducers; this was done as an experiment to see which type is most satisfactory. There was room on the circuit board to mount a Star Micronics CMB-12 buzzer, whose cantilever armature causes the housing to vibrate quite noticeably. The alternative output is a 200 Hz power oscillator fed to a small loudspeaker whose cone can be touched. (The oscillator's output is sufficient to drive a low-impedance earphone with a fairly loud signal, so that the tester can be used as an audible instrument as well.)
Because zinc-air cells have a pronounced ability to recover for short bursts of energy even after they can be declared exhausted, a simple "go/no-go" indication is not sufficient to assess a cell's condition. For this reason, a pulsating indication whose duty cycle depends on cell voltage was chosen; the higher the cell voltage, the shorter the bursts. An exhausted cell which temporarily looks fresh will, over a short number of pulsations, cause the tester to elicit longer and longer buzzes, thus providing the visually impaired user the same degree of subtlety of indication as a sighted user can get by watching the needle on the meter.
Comments from the person who received our first unit were favorable. Because of its portability, and since indications from the on-board buzzer were deemed sufficient, the loudspeaker was deemed to be superfluous. The privately funded Rehabilitation Engineering Service of Smith-Kettlewell is making the device available individually for $125 each.
As nonvolatile memory devices have expanded in capacity, and as they have begun to use CMOS technology that affords low battery drain, limited-vocabulary instruments containing commonly used verbal messages which can be operated by speech-impaired users (such as people with cerebral palsy, or profound hearing impairment) have become commonplace. Surveying the market for such devices suitable for deaf-blind, nonvocal people, however, we found that most had no way of terminating the speech once a message has been initiated (perhaps accidentally), while some will not allow the user to override his choice of message (once initiated) to select another.
Our experiments with Information Storage Devices (ISD) speech-recorder/player chips led us to the conclusion that we could design and publish a very economical build-it-yourself limited-vocabulary device with the necessary "abort" features as an integral part of the design. The resulting instrument has no external controls besides the ten message buttons. Internally, an on-board microphone, a high-level record input, and a "record" switch enables the person loading the messages to do so either from a high-quality tape recording or via the internal mike. Any wrong button pressed can immediately be overridden by a successive selection. Moreover, one or more of the buttons can be loaded with an "end-of-message" pulse; merely tapping this button will shut down the device, while lingering on this button causes the message stored in that position to be played.)
The flexibility and simplicity of this design allow fabrication in modestly equipped facilities (such as in high school electronics shops). The cost of the ISD chip and other components is well under $50.
Our multidisciplinary approach to rehabilitation research of those with multiple sensory losses is aimed at developing innovative methods to refine diagnosis, predict developmental anomalies, and devise optimal rehabilitative strategies for the multifaceted problems of vision, hearing, and combined vision/hearing disabilities. These activities, while under the umbrella of the RERC, have been carried out with supplementary funds from the Smith-Kettlewell Eye Research Institute, the National Institutes of Health, and NIDRR Field-Initiated grant support.
1. Objective Binaural Hearing Assessment
The aim of one project in this category was to develop an objective method of monitoring each channel of a subject's binaural hearing system -- a method applicable to adults, infants and localization-impaired persons. The system was intended to measure asymmetrical hearing impairments. Using evoked auditory potentials, we have determined the interaural time difference (ITD) and interaural intensity difference (IID) to within 500 Ês. With changes in methodology, the resolution might be increased to 100 to 200 Ês for ITD and 3 to 5 dB for IID. This order of resolution was not sufficient for practical utilization in an aid for localization-impaired individuals, so our emphasis was changed to psychophysical approaches instead.
2. Innovative Binaural Hearing Aid Design
A second project, supported by NIDRR Field-Initiated funds and NIH, was pursued by Helen J. Simon, Ph.D., Smith-Kettlewell Research Audiologist. The goal of this project has been to develop an innovative type of binaural hearing aid. Recent developments have focussed on developing strategies to optimize the fitting parameters for such an aid. aid. Since sensorineural hearing loss (SNHL) and blindness may interfere with localization of potential hazards, a second goal of this project is to explore and develop the parameters for improved localization utilizing a new rationale. This new hearing aid concept may permit people with SNHL and blindness, using acoustic cues, to locate and avoid hazards. Details of the project are available from Dr Simon.