Dr. Ross on Hearing Loss
Supplement your Hearing Aids: Hearing Assistive Technology (HAT)
by Mark Ross, Ph.D.
Each year as I wander through the exhibit hall at the national convention of the Hearing Loss Association of America (HLAA), I am again struck by the enormous variety of hearing assistive technologies (HAT) available (other than hearing aids). While there is no question about the necessity of a hearing aid for most hearing-impaired people, often, indeed I would go so far to say "usually" a hearing aid itself is not enough. There are many occasions when a person with a hearing loss requires some form of hearing assistance that hearing aids are not designed to provide or are not adequate for the situation.
I think about this every time I attend a large area presentation, like the yearly research symposium. I make a point of comparing how well I can understand a speaker, with and without using the available assistive listening device (ALD). While the ALDs used at the convention were infra-red systems, the listening enhancement experienced would be true with any ALD (but more conveniently, in my estimation, with an induction loop than other types that require a separate receiver). The improvement in relaxed comprehension was simply astounding, to the point where I would not now attend any large area performance or presentation of any kind if an ALD was not available. Using just hearing aids would not do the job for me. And this is what hearing assistive technologies (HAT) are all about: they supplement the contributions made by hearing aids, extending their use (as when an ALD is used) or by providing a substitute channel (for example, by converting sound to a visual or vibratory stimulus).
It is these aspects of HAT that will be emphasized in this article. In it, I will describe just a few of the many that are available; it should be understood that there are many others, tried and true, that can play an important role in easing the impact of a hearing loss. The sampling below is not meant to be comprehensive.
I was very pleased to see the Companion Mics system introduced this year by Etymotic Research Inc. It is a concept of communication access that the company has been working on for a number of years. Companion Mics are FM systems and thus offer the same advantages that these systems generally offer (a major increase in the speech to noise ratio because of microphone positioning close to a talker' mouth). But instead of a single transmitter and one or more multiple receivers, as is true with the usual personal FM system, the Companion Mics system employs multiple transmitters (up to four) and a single receiver (though there could be more). The transmitter ("talker" units) is suspended around each talker's neck and transmits their voices to the single receiver ("listener") unit. This would not be possible with current FM systems, because of the interference produced when two or more transmitters attempt to utilize a single transmitting frequency. Because of the nature of its technical design (employing a 2.4 GHz spread-spectrum frequency hopping system), Companion Mics can accommodate up to four talker units without mutual interference. A picture of the Companion Mics is shown in Figure 1. Since both the transmitters and the receivers look exactly alike (though function quite differently), only one unit is shown.
I've had an opportunity to try the Companion Mics for the past several months, mainly in two types of situations that ordinarily present a good deal of listening difficulty for someone with a moderate or severe hearing loss. The first is having dinner with three other people in a noisy restaurant. While I can usually hear the person sitting alongside me with just my hearing aids, or one across the table by using my personal FM system, there is no way that I can communicate effectively with all three people as they take turns talking. With the Companion Mics system, this was possible since each person used a separate transmitter (still, one talker at a time please!). One time it was my normal hearing companions who asked that the "music" emanating from a loudspeaker above our head be reduced! It didn't bother me one bit.
The system did not work quite as well in a situation where a group of eight people gathered around a table in a noisy restaurant. I gave the Companion Mics to three participants who I particularly wanted to talk to at that time. During the course of the lunch, however, these people would be talking to other people, while someone else was trying to talk to me, all at the same time. Needless to say, this made for a rather difficult listening situation. Right now, I would recommend the system primarily when each talker can use a transmitter. Currently, the system will accommodate only four talkers. Evidently, the nature of the transmitting mode makes it unlikely that more talker units can be accommodated using this technology. Still, four talker units do encompass many listening situations (and does encourage me to go out to dinner with my wife and friends more often).
The second type of situation that ordinarily poses great difficulty for someone with a hearing loss is being able to hear two or three other people in a moving vehicle. There are many occasions when I am in a car with my wife and another couple and I find it next to impossible to participate in the ongoing conversation. If I'm driving or sitting in front, I can't hear the people in the back. If I'm in the back, then I can't hear the people in front. Communication choices are limited in these types of situation. One can dominate the proceedings and do all the talking, attend only to one's partner sitting alongside and try to figure out at least the general trend of the conversation, or one can simply tune-out. None of these are very satisfactory options.
Now, when I enter a vehicle with two or three other people, I first place the Companion Mics around their necks and get the system going (there is a brief calibration sequence to this). And that's all. From this point on, no matter where I am in the car (front or back), I can participate fully in the discussion. This is another example of where a hearing aid alone is often insufficient, though certainly necessary. One caution, however; some automobiles apparently generate a great deal of ambient electro-magnetic energy that can completely mask the speech signals (this happened to me when I was a passenger in a new Volvo).
The system comes with insert earphones which, however, proved to be inadequate (and not convenient) for me. Instead I use a neck loop which works quite well. There is a simple charger into which both the talker and listening units are placed. Battery life is estimated at 10-13 hours. Each unit includes a volume control, with the listener's unit serving as a master volume control. While the unit is specifically designed to accommodate multiple talkers, it can also be used in any situation in which a conventional personal FM system is ordinarily employed, that is one on one. Further information about the Companion Mics can be obtained from www.etymotic.com
Various varieties of telephones seemed to be the most frequent type of device displayed at the convention. The tables in the exhibit area were full of them and the catalogues listed even more. This emphasis is understandable considering the essential role that telephones play in our lives; it's hard to image our living a full life without access to telephone communication. Hearing-impaired people, too, require an effective way of using the telephone. Listed below are the most popular options and telephone features now available. (Purely visual systems, such as the varieties of TTYs, are not included here.)
The system works by the automatic routing the CapTel user's telephone call through a captioning service. In a single-line system, the user calls his or her party in the usual way. The call is automatically directed to the captioning service where a captioner, using an advanced word recognition program, transcribes their words into captions that appear on a small screen on the user's telephone. People with visual problems may elect to route the captions to a computer screen via a USB port and enlarge the captions to the extent desired. There is evidently a small delay (perhaps two or three seconds) between a caller's utterance of a word and its appearance on the user's screen. A user can decide whether to display the captions or not.
In order for an incoming call to be captioned, however, a caller first has to call (1 877 243-2823) and then enter the destination's phone number. There are separate call-in numbers for Federal Relay users, Spanish-to-Spanish users, and California residents. If one of these numbers is not dialed first, then the call is not captioned. This can be a problem if one requires, as one should, full access to all incoming telephone calls.
This is taken care of in a two-line CapTel service, in which all incoming calls are automatically captioned. The phone itself then routes everything the caller says to the captioning service, which then captions the call and directs it back to the phone via the second line (albeit with a slight delay). In other words, the captions are carried on one line, while the acoustic connection is carried on the other. The extra cost for the second line is determined by the local provider; no extra charges are applied for the captioning service itself.
According to a representative of the Ultratec company, they obtain a recognition accuracy of 98% or higher with their proprietary word recognition program. Each operator receives a lengthy and extensive training program to prove that their speech can be converted to text with this degree of accuracy. This is an extraordinary accomplishment and I know of no independent confirmation of the claim, though I have no reason to doubt it. The captioning service is under the same requirements as any other Telephone Relay service; there must be someone available 24/7 for 365 days a year. Right now, CapTel is available in 39 states, but evidently other states are also interested in providing the service. Further information about the CapTel phone itself can be obtained from ultratec.com, or captionedtelephone.com. To find out how to obtain a phone, call 1 800 233-9130.
Dragon Naturally Speaking, version #9
Dragon Naturally Speaking is an automatic speech recognition (ASR) program, much like the one used in the CapTel phones. What any such program does is convert the spoken word to print. While such programs have been around for a while, and improving all the time, the general consensus had been that they were not for "prime time", that is unrestricted speech by an average talker. Talkers had to undergo a lengthy training program in order to familiarize the software with the unique acoustic characteristics of their speech, the recognition vocabulary was insufficient for many applications, and homonyms of words would appear rather than the one intended; the system could not differentiate between homonyms on the basis of the linguistic context. Non-fluent speakers, the ones who fill their speech with all sorts of irrelevant sounds, whose speech seems to meander in an endless maze, seemed even more unintelligible with an ASR system. Available ASR systems appear most useful under conditions where a familiar speaker dictated a report or narrative on some restricted topic, or where word choices are limited (such as when making reservations on the telephone).
Evidently, at least according to the report I read in the July 20th issue of the NY Times, the latest version (#9) of Dragon Naturally Speaking, is a highly accurate word recognition program that does not require a speaker to train the software by reading pre-set prose material. In the author's personal test, he read a 1000 word excerpt from a book and found that the software got only 11 words wrong. This is a 98.9 percent accuracy level. He did even better (99.4 percent) accuracy when he trained the software by reading an Alice in Wonderland excerpt. Also, the program evidently uses context to differentiate between homonyms (for example, between bare" versus "bear."). This is a phenomenal score to achieve with an ASR.
I do wish that we could accept these results uncritically. It would such a boon for people with hearing loss. Imagine watching live TV and actually understanding what a speaker is saying via the captions! Right now, trying to understand such TV programs is often quite frustrating and sometimes impossible. (I am not talking about pre-captioned programs, which are great). With not too much imagination, we can see numerous other possibilities that a 99.4% accurate word recognition program would have for people with hearing loss. Imagine, for example, reading conversation in real-time on a hand-held device! What this feature article in the NY Times does not cover is how the system would work in a conversational setting with an unskilled talker, and not when an articulate speaker reads from a book in a controlled situation. So we do need to interpret such reports with a degree of caution. At the same time, however, the progress in ASR programs that is clearly taking place is quite heartening. Further information about Dragon Naturally Speaking can be found at nuance.com
Several Miscellaneous Devices
Handheld Captioner with FM Receiver. Naturally, at a convention devoted to people with hearing loss in a place like Orlando, one would expect that the availability of communication access at Disneyworld would be a topic of interest. To their credit, dating at least from an earlier convention in Orlando, the company has made communication access a priority consideration. Presently, they provide various kinds of assistive devices at their attractions, ranging from reflective captioning to both FM and Infra-Red listening devices. Their newest device, still being worked on, was exhibited at the convention. It is a handheld captioning unit which displays narrative text at locations in which fixed captioning is impractical, such as a moving attraction. The receiver is no larger than blackberry and can be easily carried. It includes an FM receiver so that the user can both hear and read the narrative. So the next time you go for a ride to the moon from Disneyworld, you'll know what the spaceship captain is saying!
Portable Induction Loop and P.A. System. Portable induction loop systems and Public Address systems have, of course, been around for years. What I found neat about this device was the availability of both amplified sound and a loop signal in a single unit, as well as the inclusion of a separate wireless microphone. The unit attracted my interest after my experience last year when I addressed a group of mainly hearing-impaired people. The only communication assistance available was a CART reporter, which while commendable, was less relevant than audition for a number of the attendees (normal and hearing-impaired). This portable system would have provided communication access for both groups, either with the P.A. system or via the loop system coupled to a telecoil in personal hearing aids. And, evidently, at a reasonable price ($400 at the convention and $450 thereafter). Furthermore, since it also includes a wireless microphone, it can facilitate audience participation by a group of primarily hearing-impaired people. Further information can be obtained through HARC.com or by calling HARC Mercantile at 1 800 445-9968.
Receptive Communication Needs Profile
I began this article by noting the enormous variety of hearing assistive technologies that are available. Most people with hearing losses are not fully aware of their diversity and function. It is also pretty safe to assume that many hearing aid dispensers are equally uninformed of the full range of such devices. In the usual hearing aid selection scenario, an audiologist will inquire about areas in which a client experiences particular communication difficulty. At that time, but in the course of a primary focus on hearing aids, the audiologist may mention the existence of common some assistive devices that he or she is aware of. Maybe there are some on display in the waiting room and maybe not. Maybe the audiologist (or some aide) will take the time to explain the existence and purpose of some device and maybe not. Or maybe the client will simply be referred to a Radio Shack store or maybe not. My point is that the selection, dispensing and training in the use of HAT are a bit of a hit or miss function within contemporary hearing aid dispensing practices (with a few notable exceptions here and there). Too often, the need for some HAT is completely ignored during the hearing aid selection process, with the responsibility for knowing about and acquiring such devices left completely up to the person. In my opinion, considering the importance of HAT for many hearing-impaired people, these kinds of practices are simply inexcusable. It deprives hearing-impaired people of possible additional auditory assistance that can have a major impact on the quality of their lives.
But this leaves the question of how one goes about evaluating the full range of a person's communication needs, beyond the invaluable assistance that hearing aids can offer. One would expect that people with hearing loss would be able to pinpoint their own areas of hearing difficulty and thus narrow the search for a particular hearing assistive device. But this does not happen in real life. How many people with hearing loss would be aware of the full range of telephone features reviewed above? Or know of visual and vibratory smoke and wake-up alarms, the many types of TV listening systems, or the latest development in personal FM systems? A systematic interview by an audiologist can elicit many of these needs, but others will remain unstated and thus unserved.
There is, in brief, a need for some sort of navigator, one that could identify and then point to particular devices that are specifically targeted to a person's communication needs. While some audiologists have locally developed questionnaire forms for this purpose, no such instrument is currently available that has been validated on people with hearing loss and that employs modern computer technology to aide in the selection and identification of particular devices. This is the intent of a project now being carried at by Dr. Cindy Compton Conley at the Rehabilitation Engineering Research Center (RERC) at Gallaudet University. The goal is to develop a software program that can assist audiologists in analyzing an individual's receptive communication needs and then identifying particular assistive technology that can meet those needs. The program will not only highlight specific devices (and vendors), but will also suggest other aural rehabilitation procedures appropriate for that person. Stay tuned for progress reports!