Assistive Listening System Project                                                                          Lexington Center RERC

October 7, 1999

MR. BAKKE: My clock says it's 10 o'clock.

I'd like to welcome all of you here.

What I want to do first -- I'm sure there are going to be people drifting in as we move forward, that's okay.

We'll take time, during the program, to introduce people.

What I'd like to do, first, though, is introduce our crack staff of researchers.

Most of you know Mark Ross, who is senior researcher with the RERC and an elder statesman of audiology, and a heck of a nice guy.

Faye Erickson is another researcher with us. Faye is going to be reporting on the research that we have done on listening, listening standards for assistive listening devices.

Mauro Caputi is another researcher with us, an electrical engineer -- Mauro is an electrical engineer, Hofstra University, and works with us two days a week.

We have Mike Steele, who is another electrical engineer, full-timer with us, knows a lot about directional microphones, because we've been working on a project developing a directional conference microphone system.

Deirdre Walters, research audiologist, who has been with us, what, about three years.

We have Liz, her last name is -- Polski. Liz is an NIDRR scholar. We receive money from NIDRR to help students learn about research and hearing enhancement. She's been with us about a year.

And we have Luciana, I want to say Pavarotti, but that's not right. She's a student at St. John's University.

Sorry, Liz is a student at NYU and Luciana at St. Johns University.

That's my senior...

You've been with us about, almost a year, right?

Yeah, almost a year.

Harry Levitt will be here shortly. He's coming in from New Jersey.

Did I miss any staff?

Of course we have Don DePew, who is taking the notes for us. Don is well-known to the hard of hearing and deaf community in New York City and is an excellent recorder.

And now what I'd like to do is pass the mike and have each of you introduce yourselves, so we know you. Just kind of say your name and what organization you're from, and anything else you want to say, as long as you say it within 30 seconds?

MS. SORKIN: Hi, I'm Donna Sorkin, actually here wearing two hats. I'm here as a representative of the Access Board, the architectural and transportation compliance board. I'm a board member. The board is the entity that actually commissioned this research.

My other hat, and the one that's mentioned in the program materials, I'm the executive director at the Alexander Graham Bell Association for the Deaf and Hard of Hearing. "And hard of hearing" was added in April.

So we're primarily an organization that focuses on the needs of children. Our membership has parents, professionals who serve kids, and I have about 10 to 12 percent of our members are adults with hearing loss.

I'm very glad to be here.

MR. BAKKE: Just noted, Dr. Cohen walked in. Oscar Cohen is the superintendent of the Lexington School for the Deaf and also the director of the Lexington Center for the Deaf.

Dr. Cohen: Hello!

I'd like to welcome all of you here, on behalf of Lexington, our trustees and our faculty and staff.

This is an important meeting. And the longer I talk the less work you're going to get done. So have a productive day. I'll be very interested to see what your outcomes are.

Again, welcome to Lexington.

Be careful if you're going to Manhattan later, because the president is in New York. He is attending a dinner and speaking, for the first time since about the last six presidents or five presidents, at a Teamsters dinner for James Hoffa. So just watch the traffic, if you're going to Manhattan.

MS. HO: Yvonne Ho, representing Phonic Ear, which is a manufacturer for assistive listening devices.

MS. STIERNBERG: Jeanne Stiernberg, independent consultant in the electronics, industry, also a speech-language pathologist, here representing two of my clients, as well as just having a personal interest in this.

One client is the National Systems Contractors Association, and another client is a publisher, Miller-Friedman, that publishes System Contractors News and Pro Sound News.

MR. GENDEL: I'm Josh Gendel, director of the Assistive Technology Center at the League for the Hard of Hearing.

MR. GLICK: I'm Ira Glick, President of Sound Remedy, which is a very small organization that provides consultation and services for sites mandated for hearing access by the ADA.

I just had a question, Matt. Considering the subject and the people here, is this room set up for assistive listening?

Is it looped or what?

MR. BAKKE: It's looped.

MS. GLICK: I'm Paula Brown Glick. I'm on the cultural affairs committee of Advocates for Better Communication, which is an organization for awareness, access and advocacy for people who are hard of hearing.

MR. MEYER: Hi, I'm Eric Meyer. I work for Sennheiser, manufacturer of assistive listening devices.

MR. DAVIS: Good morning, I'm Bruce Davis, with Gentner Communications. We manufacture assistive listening devices for the hearing impaired.

MR. YUAN: I work for David and Dr. Matt.

I also work for Michael.

I'm working -- multi-directional microphones, and that's my job.

MR. BAKKE: Sorry, River wasn't here, when I was introducing the staff. River is an electrical engineer with us, and a Ph.D., Dr. River.

And this is Leroy Weeks, who was not here earlier. Leroy is a technical engineer with us.

And Harry Levitt has come in.

DR. LEVITT: Hello, I'm glad to be here. I'm glad you're all here.

MR. BAKKE: We'll move on.

MR. CAMIN: Good morning, my name is Fred Camin, Brooklyn chapter of SHHH. The president and the director of the devices -- and I'm here as a consumer.

MS. GINSBERG: Good morning, Judy Ginsberg. I'm here from New Jersey. I represent SHHH State Association of New Jersey. And in the future, the Director of the Division of the Deaf and Hard of Hearing in New Jersey is planning to open up an assistive listening device center and I will be in charge of that.

MR. GORDON: Good morning, my name is Joe Gordon. I wear two hats. I'm a member of SHHH and the New York State chapter coordinator for SHHH.

I'm also a member of Advocates for Better Communication, which is allied with the League for the Hard of Hearing in New York. And at the Advocates for Better Communication, I am the chair of the telecommunications committee.

Thank you.

MR. McGIBBON: Good morning, my name is Ed McGibbon, vice president of the Manhattan chapter, also a member of the Westchester chapter, involved with many issues with hearing loss. Also on the Board of the New York State association and I'm looking forward to hearing more about listening devices.

MS. SONNENFELD: I'm Myra Sonnenfeld, founder and, again, President of SHHH South Nassau Chapter. We have the largest assistive devices collection, I believe in the United States, bigger than SHHH national or the League for the Hard of Hearing. And we help helped start the one at Lexington.

I'm a retired teacher of the hearing impaired.

MR. JOHNSTONE: I'm Charlie Johnstone, people call me Chaz, audio visual, at the National Technical Institute for the Deaf, manage all the assistive devices, TTYs and anything else people don't know what to do with.

MR. ROSS: And Matt introduced all of us, so...

MR. BAKKE: We had one other person come in: Anita Haravon, a researcher with the Rehab Engineering Research Center here at Lexington, doing work with computer speech recognition systems.

I just want to, very, very briefly, give a little bit of background before we move into the research report.

Again, this program was begun in response to a request by the United States Access Board, to help us develop standards. To help them -- for us to help them develop standards for assistive listening systems in public areas.

This was a consumer-driven effort. I think Donna had a lot to do with it. So we have a consumer-orientation. So we'll say that right at the outset, that we do have a consumer orientation.

The purpose of the several meetings that we've had, is to bring the consumer concerns to the fore and discuss those with the people who manufacture and distribute these systems, and install these systems, so that we can find common ground and really move forward and come up with some recommendations that make sense.

The research objectives of this project were to review the state-of-the-art. We did distribute a state-of-the-art paper that was written by Mark Ross. All of you should have seen that.

We also needed to survey consumers, service providers, dispensers and manufacturers, and to develop an objective means of specifying the characteristics of an assistive listening device, in order to predict how well it would work in practice.

And finally, of course, dissemination.

You'll be pleased to know we have received some more funds from the Access Board for the purpose of developing some dissemination devices we'll begin working on in December.

So what did we do?

We had a consumer group right in this room -- about two years ago, almost now -- in which we gathered a great deal of information from consumers. And then we had another focus group in October of the same year, with many of the same people who are in this room now. And we took that information, along with other information, and prepared a state-of-the-art review.

At the same time, we began the development of an inexpensive measure of what we call the Speech Transmission Index. This is an index that enables us to make predictions about how well people are going to understand speech in reverberation and noise. And you'll see in the research that the STI is very helpful in making these predictions.

We also needed to look at the relationship between the Speech Transmission Index and other variables, and tolerable limits of interference for hard of hearing listeners, and that is the research that we'll be reporting this morning.

And of course, dissemination of results will consist of a report to the U.S. Access Board and, again, some follow-up activities.

The categories of concerns that we heard from consumers -- I'm not going to go into detail on these. We've done this several times and it appears in the paper, but there were electroacoustical concerns, in terms of how these systems sound and the consistency of these systems across various venues.

There were installation-related concerns, again related to how these systems are installed. The installation of these systems is critical to what the outcome is, in terms of the listener satisfaction.

There were logistics-related concerns that we'll be talking about today. And these involve issues of distribution of the systems, advertising of the systems, signage and that sort of thing, which, again, advertising and signage-related probably should have been subsumed under "Logistics," so those were the kind of general areas of concerns that we heard from consumers.

And from manufacturers, installers and service providers, we heard a whole range of concerns that are listed there. I won't read them.

All of these things, we're going to be discussing today. And the outcome of this meeting should be some tools that we can used to make our recommendations to the Access Board.

We are not making the standards.

What we're doing, is we are making recommendations to the Access Board. Because it is the Access Board who will make the decisions about the standards. But this is our opportunity to have a huge voice in what eventually comes out.

Now, in terms of the signal quality, I just want to say, very briefly, something about the kinds of recommendations we're making with regards to signal quality.

This is the model that we used initially to talk or to think about assistive listening devices and what kinds of complexities are involved. And as you can see, it's a very complex picture.

There are many possibilities in terms of sound source, acoustical, electrical sound source. It could come from a live speaker like we have now; it could come from a soundtrack.

The input or the output from the transmitter could consist of an acoustic output, like a loudspeaker, electrical, magnetic, radio, FM, infrared.

And it also could be by-passed and go right to the assistive listening device and there were several channels or media that could be used for the input and several other media for the output.

The personal amplifier, several channels are possible on the input and two on the output, either acoustical or mechanical.

Again, with the listener, receive acoustic input or mechanical, and actually we should have said an electrical input in the case of a cochlear implant.

So it's a very complex picture. So how do we resolve that?

We're suggesting that the standards address signal quality at the ear of the listener, and that's the approach that we've taken. It's up for debate, if anyone disagrees with it.

Furthermore, for microphone inputs the evaluation must take into account the acoustical space between the microphone and the sound source, because of issues of reverberation and added noise. So that is the approach we've taken and that's the rationale.

The other area that we want to be talking about today, later in the day -- after lunch, I believe -- is the issue of installers.

To us this, I think, is a critical issue, since the skill of the installer in putting these systems into practice is a crucial variable. You can have the best system in the world: If it's not installed properly, you're going to have huge problems.

Okay.

Now, at this point I'd like to just turn the mike over to Mark for a minute, Mark is our elder statesman. He's going to make some ground rules for the discussion and then we're going to go right into the research report.

MR. ROSS: Matt, you keep saying "elder statesman," I think I should start walking off now with a crutch.

Just a few ground rules: We do want to stay on topic. So I will ask, later on, when anybody talks, you first state your name so Don will know who's speaking and make your comments to the point.

Please, no speeches. We do want to hear everybody, but we do want it germane to the topic.

We want to also acknowledge the fact that this is a consumer-driven project. It came out of, actually, I have to say some advocation from Donna as a member of the advisory group for the Access Board, who received a lot of complaints from consumers.

All of us here who are consumers understand the problems that these systems -- that these systems give us.

So what we want to do, then, is try to come up with recommendations that the Access Board can develop standards which will be in the benefit of consumers.

So the ground rules, basically, are simply this: When you talk, please state your name and make your comments. And we want to hear from everybody. Make your comments to the point.

Do I have any other ground rules, as the elder statesman?

I am looking forward to this. I think this is the culmination of several years of work. And it's one of the important projects, I think, that the RERC is engaged in that will have an impact, we hope, throughout the United States.

MR. BAKKE: Just, I forgot to mention that the room is looped, so anyone --

MR. ROSS: For those of you wearing hearing aids -- which I think represent a good majority of us, and I'm sure those of us wearing hearing aids have T-coils, right -- the room is looped. So you can access with the loop.

Can you get it, Donna, through your system?

Yeah, okay.

So if we don't have access here, in this kind of meeting, we can just give up!

So we have access.

MR. BAKKE: We have several other devices, if anyone needs them.

MR. ROSS: We have several other devices. We have some FM receivers. If anybody needs it, we can give them an FM receiver.

I'll be talking to you later this afternoon, but right now I want to turn the meeting over to Faye, who is going to present the research.

MS. ERICKSON: Thanks, Mark.

Good morning, everybody.

Since we began this morning, Mary Rose has walked in. She was involved in this project. Approximately half the data that was collected at Hackensack University Medical Center, which is where Mary Rose is the director of audiology there. The rest of the people on this list are people you have met now, and they all participated in helping to get the research.

Sorry, I'm used to having it attached to my clothing, not movie star things.

Anyway, this is the title of our study, as Matt mentioned, trying to come up with some standards, quote unquote, electroacoustical standards for assistive listening devices.

We were trying to establish guidelines for specifying acceptable output characteristics of assistive listening devices for people with hearing loss. The way we went about doing that, was gathering quality rating judgments for the people. The people who participated in our study did not tell us what they heard, told us what they felt about what they heard.

The stimuli we chose to use for the study are what we call the Lexington dialogue sentences. And it's a bunch of sentences, stimuli that we created a number of years ago, which consist of a man and women engaging in dialogue. From this pool of sentences we selected six people who, of course, were not having to tell us what they heard. They got to tell us what they thought about what they heard, because they heard these sentences over and over and over again under various different kinds of distortion.

So they just -- these were the sentences that we selected. Nothing particularly special about these sentences, just that half of them were spoken by a female talker and half were spoken by a male talker, same female and male talker throughout.

We chose to look at three different kinds of distortion. It's a little confusing, because we have five different conditions. The first type of distortion we were looking at is room reverberation distortion. With reverberation, since we don't have a reverberant or -- what are those rooms called -- the antireverb room, since we don't have one of those you have to include noise in there as well, reverberation plus noise. Since there was no way of getting rid of the noise, for the reverberation plus noise condition, three separate rooms we selected. We selected the auditorium right here in the school. There we were able to get pretty faraway listening conditions.

We selected the conference room, which is directly across the hall. For the conference room, I think you should take the parentheses off the noise there, because it was definitely a reverberation plus noise room. It had this type of system, this heating system going on and it was very, very noisy.

And the next room that we selected was a classroom, a typical classroom. Again, we had different distances in that classroom.

And I'll explain what I mean about the different distances. I'll go into the recording conditions and how the stimuli were actually prepared and how people actually listened.

Recorded noise coming from an FM loop system. I think I can play that for you. Unfortunately, I imagine a lot of you are familiar with it, have heard it in the past.

What we did with the FM loop noise, we had five different signal-to-noise ratios. People listened to the speech mixed with the noise in five different signal-to-noise ratios.

It's on the floppy there.

It should be right up there.

MR. BAKKE: It has to come alive.

MS. ERICKSON: While he's setting that up...

And we also did peak clipping. So six different levels of peak clipping. Again, I'll describe how we did that in a couple of minutes, but these are the conditions that we selected.

I play it right off the floppy. It works.

So for the classroom -- I'll come back to this slide. I'll describe how we made the recording conditions.

For the reverberation plus noise conditions, how we set it up, is we made the recordings onto a digital audio tape.

(Background noise tape blasts out)

Whoa!

Digital audio tape recorder, we had these sentences, quiet sentences that we played from a loudspeaker. The loudspeaker in this case was an artificial mouth, BNK artificial mouth. Looks very unusual, supposed to simulate the radiation characteristics of a human vocal track.

How we played the speech coming from that, trying to simulate a real talker: The recordings that were then made using a sound level meter microphone, high quality microphone, the output from the sound level meter then directed to a digital audio tape recorder.

The way we would have it set up, the digital audio tape recorder and sound level meter sitting in one position, then the speaker simulating the microphone -- not the microphone, the speaker. The artificial mouth was then moved away at successive distances from the recording device.

Okay?

So the sound -- the listener, in this case -- the sound level meter was always in a fixed position.

Immediately following the presentation of every sentence, each recording of every sentence, did a Speech Transition Index at each distance away from the microphone. Record the sentence, then do STI; record the sentence, then do STI.

We have been involved in developing a Speech Transmission Index program ourselves here. Didn't use that one, because it was slow. I understand it's been speeding up a lot.

For those of you who are not familiar with Speech Transmission Index, want me to give a brief background on it: I'm not going to do it. I'm going to let Harry give a brief background on the Speech Transmission Index.

DR. LEVITT: The words to an old song say "the song has ended, but the melody lingers on," that's really what happens with reverberation: The acoustic stimulus ends, but the reverberation continues.

That's the crucial aspect of reverberation: The ongoing part after the sound has ended. The technique known as the Speech Transmission Index, looks at the continuing component of the reverberation after the sound has ended.

There is a large body of data on the effect of noise on speech. And so it transforms that reverberation continuing sound into an equivalent noise. From that equivalent noise we can predict the intelligibility of the speech signal, and that's the basis for the Speech Transmission Index.

MS. ERICKSON: Thank you.

To give you an indication of how the Speech Transmission Index changes at the different recording distances, does that look pretty clear?

What this shows here, is the blue symbol for the classroom condition. Those were fairly close distances that you would have in a classroom, relative to the auditorium.

What this shows is that the STI are very close. 3-inches away was, of course, very high. As you got progressively further and further away, the STI got poorer and poorer. This little line along here, what this is: In the classroom when I first was using this software, I wanted to feel confident it was doing a good job, so I measured every 3-inches. So that line is just showing what every 3-inches, the measurements at every 3-inches. But along there I wasn't going to bore the listener. I just selected points along that function.

And then these, the auditorium, again further distances. I think we went up to about 10 feet, I think, in the auditorium, maybe further.

No.

No, 15 feet, 15 feet in the auditorium. Again, five different locations in the auditorium that we made recordings. It just happened to work out nicely. We could fill in the gaps there on the function for the classroom.

Then the bottom one is the conference room and that is really highly influenced by the background noise. So the STI in this case is poor regardless, because of the background noise. This hissing sound going on in the background, the distances there. The way we made the recordings in the conference room -- one thing that is also an influencing factor, not just the hissing sound from the heating system, also we had the microphone in a position that a listener would sit in at a table, so you also had reverberation coming off of the table as well.

So, a nice reverberant source right there. We always had very important STI scores for the auditorium.

Okay, this is the perfect time to bring up the loop noise. This is a spectrum of the loop noise. Matt is going to play the loop noise.

MR. BAKKE: That's not it.

MS. ERICKSON: That sounds like it.

This is a spectrum of the loop noise that we used.

(Playing loop noise)

MS. ERICKSON: We heard enough of that already, I'm sure.

What we did with the loop noise, is we then digitally combined that with the speech. So we had different signal-to-noise ratios varying from 20 to 30.

Yes, 30 dB in 60 successive steps, both for the male talker and female talker. This is not working as well as I thought. When you put these things on, always get that electrostatic thing that causes them to move, no matter what you do.

You guys kind of have the idea: The speech and the noise.

Here is just a spectrum of the male talker, his speech. A lot more low frequency energy than the female talker. The female talker has very high frequency energy in this case, which happens to be me, so...

And the last condition that we had was the peak clipping condition.

And there's the original wave form at the top and there's a picture of a clipped wave form, in this case 18 dB of clipping. Clipped a wave form at 60 dB, success of, no clipping, 6, 12, 18, down to 36 dB of clipping. So it was pretty well clipped at the end.

That's just a little blown up picture of it, to give you a better idea of what the clipped wave form looks like.

So this is just a review of the five different conditions that we had. The conference room, from 36, which is 3 feet, to 5 and a half feet, to 6 feet -- 4 and a half feet, rather, to 6 feet, to 7 and a half feet, to 9 feet, is we did it in successive steps, 18-inch steps.

Auditorium, out to 15 feet.

And the classroom, fairly close distances, the longest 10 feet, and the peak clipping, again in 6 dB successive steps, and the loop noise, again in 6 dB successive steps.

Fifty-seven listeners that participated. Ten of them had no hearing loss. And the rest of them we divided up into four different groups according to their degree and configuration of hearing loss.

The first group, group No. 1, we called them mild hearing loss group and there were nine people who were in this group. They all had to have pure tone averages less than 41 dB, to be considered in the mild hearing loss --

MR. BAKKE: In the better ear.

MS. ERICKSON: Yes, in the better ear. We did have in one group, a couple of unilateral people, so...

We still included them, two unilateral, maybe three.

So that's group No. 1's audiogram, these spots. Don't get too concerned, just represent ears, a lot of overlap. If you count them, they don't add up, a lot of overlap on thresholds.

And this is a moderate group, moderate to moderately severe. This is with a pure tone average greater than 41 dB, but you had to have a threshold at 2,000-hertz less than 70, a fairly -- a flatter configuration, threshold grade greater than 70, poorer ear for -- two greater than 70 dB at 2K. For those of you who are looking closely at this group, I looked at it myself and said: Oh, no, where did they come from?

But...

And then the third group, group No. 3, is the gradually sloping losses, moderate to severe sloping losses. Again, these people had to have a pure tone average greater than 41, and had to have thresholds greater than 70 dB at 2,000-hertz.

And the last group -- and I just want to mention, too, we had a ton of people, relatively speaking, in group 2. To point that out, 20 people in group 2. All the rest of the groups seemed to have about 10 people each.

This is working data. I'm glad you guys are all here today to give comments and criticism. It's really a nice opportunity to hear from you all.

So group No. 3 -- whoops, you guys have already seen that.

Group No. 4 is the precipitous loss people, hardest ones to find for us, and there are eight people in that group. These people had the ski-slope hearing losses.

Thanks to Hackensack University Medical Center, that's where we got all of those people, I think. I called Mary Rose, said: Help! We need them.

And then the normal hearing group. There's nothing really special to see there.

What the listeners were instructed to do -- we made then stimuli from the recordings that we had prepared. Listeners instructed to give quality rating judgments. To give you an idea what they were: Listened under earphones, standard TDH 49. When you listen to the audiologist, get your hearing tested, you listen under those earphones. Instructed to find a comfortable listening level. Did not adjust each ear independently, both ears adjusted to the same level, so...

This is just a figure that shows you what the frequency response of the TDH 49 earphones were: Go out nicely and then drop off pretty rapidly thereafter.

The listeners were instructed to respond using one of these four categories, so: Excellent; good, they would purchase a ticket for the show if it sounded like this; marginal, they may or may not purchase a ticket for the show if it sounded like that; and unacceptable, definitely not, just sounded too bad, too horrible.

Numbers on the side, for calculating the statistics on this. We assigned them numerical values, so we had excellent as a 4, good as a 3, marginal as a 2 and 1 as an unacceptable.

This is to also give you an indication of the listening levels that the people selected.

With preferred listening level, this is where we adjusted the level to where they said it was comfortable for them to listen to.

Again, they weren't particularly -- not necessarily the level they had to understand everything. Just had to say it's not too loud and not too soft, not uncomfortable for them to listen. Used the quiet sounds for adjusting the comfortable listening level. Unadulterated, undistorted signals are the ones they adjusted to come up with that.

And so, fairly logically, it looks like here, that if you had a higher pure tone average you required a higher listening level. It seems like if you look at each of the groups, each group seemed to have their 20 dB range from which they wanted to listen.

So for the group No. 1, from about 75 to 95. The mild hearing loss group, kind of a 20 dB range.

Group No. 2, that was the huge group, they did not go on a 20 dB range, had more like a 40 dB range. Group 3, 20 dB range, 90 to 110. Group No. 4, from 75 to 95 again. So depending on the hearing loss category, 20 dB range seemed to cover them, except for group 2, which was kind of all over the place.

Okay.

Now, the results. Remember, I described to you before, 4 being the best, listener said No. 4 was excellent, 3 good, 2 fair, and 1 poor.

These are the listener ratings for the classroom, so the no hearing loss group along the top, then groups, 1, 2, 3, 4 along this column.

The way we have this sort of color coded for you, from 3 to 4, any ratings from 3 to 4, in black. Any ratings from 2.5 to 3 in blue. Any ratings less than 2.5, in red. Any rating less than 2.5 to be -- not really, as either a consumer or a manufacturer, something you want to go for. You don't want people to say it's marginal. You may or may not buy the ticket. You want them to say: This is a pretty good system, it sounds good. This is what we're trying to aim for here.

Looks like, from looking at this data, that you cannot have any speaker-listener distance. Anything that is greater than about 2 feet or 4 feet -- probably about 4 feet. Between 2 and 4 feet, anything beyond that, people are just saying it's all unacceptable to them.

And that distance also seems to hold for the auditorium, another environment. Where we said it was 2 to 4 feet for the classroom, well, that falls right here. 4 feet falls right between 3 and 6 feet. The distances changed in the auditorium, much further out distances than in the classroom, all red beyond 6 feet. 6 feet and beyond, unacceptable, speaker too far away from the microphone, picking up reverberation and noise.

Now, the conference room, they only liked the quiet ones. Nobody liked any of them.

What's really interesting about this data, you'll see from the next two overheads as well -- I'm sorry, this is not accurate, this top column, 3, 4.5, 6, 7.5, and 9, that's not correct along the top there. I apologize for that.

But what's interesting about this, what we found interesting: Is that we expected, originally, predicting how the data would look, expected the people with greater hearing loss to find the recordings of poorer quality than the people with lesser hearing loss.

It doesn't necessarily seem to matter. The degree of hearing loss doesn't seem to really be a factor here. People thought it was terrible, regardless of how well their ears were working, so...

I thought that was interesting.

And you'll see the same thing from the next overhead as well.

So this is for the FM loop noise and you really can't have a signal-to-noise ratio poorer than 12 dB, which I think, consistently, many studies have shown, time and time again.

What's the average classroom?

6 dB, yeah, that's the average classroom.

So you have to have it better than the average classroom, basically, for people to say that it's acceptable, whether it's 8 feet or whatever in there, we didn't have those step sizes. It has to be better than that for people to find it acceptable listening. Again, didn't really seem to matter what degree of hearing loss people had, including people without hearing loss.

Okay. And for the peak clipped speech, can't have any clipping greater than 18 dB. Which also sort of makes sense, when you think about the way speech waves form. Peaks in speech, about 12 to 18 dB. If you clip off the meat of the wave form, that's considered to be unacceptable to people. You start -- when you start doing that, so...

Okay. All right. So up until now, we've only talked about distance from the speaker-listener, having really talked about the Speech Transmission Index, measured that at each of the successive distances.

What this overhead shows, is the listener ratings on the Y here, as a function of the Speech Transmission Index.

Remember, we had our cutoff is 2.5. Anything less than 2.5 is considered to be unacceptable. And these spots along here, it gets to be too busy if I had a key, but it was the same groups that were in the previous slide. So the gold circles, I think, are the group No. 3 -- I have it written down here somewhere. Let me just...

I have it on my... yeah.

Yeah, the red squares, group No. 1; the green circle, group No. 2; and the gold diamond, group 3; and group 4, I did not include the people without hearing loss in this group -- in this figure.

What I did was put a simple line to their data.

It works out if you calculate -- if you solve for X here, works out that the Speech Transmission Index that correlates -- that agrees with this rating of 2.5, happens to be .76.

Any Speech Transmission Index that is poorer than .76 is considered to be unacceptable.

And what you would miss, if you look at this in terms of the ratings in listener, if you made it -- you're missing these people, essentially, right here.

These people still are getting poor ratings of less than 2.5, but your STI cutoff would allow for them to be listening under these poor listening conditions, what they consider to be poor listening conditions. I'm thankful that they are not way down here.

Each of these spots, remember, don't represent a group of individuals. They represent the average for a group of individual listeners, probably some above 2.5, and they found it to be acceptable.

And if we were to look at this...

If you decided you don't want to do this STI thing, seems like a lot of work, why not do distance? Get out your ruler, measure how far away it is, seems an easier way to go. Fit a line with this data. Again ratings on this axis, but inches on this axis. So the orientation of the curve changes. You have to readjust your thinking on this.

Fit a simple line to that data, it comes out that you can't be any more than 55-inches away. However, you are missing, now, this entire group.

Okay, these people still -- these listeners still said it was unacceptable, but they are within your 55-inches of acceptability, you know, criterion that you've set. They fall underneath that.

So it seems, from looking at these two figures, that STI is the way to go encompasses both signal-to-noise ratio and distance as well. And it's a fairly quick measure and I think it's a logical way to go.

So...

Anyway...

In finishing up here. Based on what we found, this study, we consider the signal-to-noise ratio cannot be less than 12 dB, which is what we talked about. The Speech Transmission Index cannot be less than .75, based on what we found. And peak clipping levels cannot be more than 18 dB. And the system must be capable of providing 110 dB SPL based on the preferred listening level, the figure I showed you a while back.

And that's basically it. It sums it all up, from what we found.

I'm ready for questions, so, please...

MR. BAKKE: I'm going to ask a question, then I'm going to answer it.

In terms of microphone distance, an important factor there is what kind of microphone you're using. Because depending on the type of microphone you're using, distance can vary.

There's something called a distance factor with a directional microphone. So you have to factor in the kind of microphone that you're using.

In this case we used a very high quality omni directional microphone.

We had done some research a couple of years ago, on conference microphone systems. And we had compared several conference microphone systems to this omni directional microphone we're talking about, which is -- it's a B&K, a Grulencar --

Which I don't expect you to spell, because I can't.

-- microphone that's used in very careful measurements of sound level. So it's a very good microphone, very good omni mike. So that's the mike we're talking about in terms of distance.

What we'd like to do -- are there any further questions about the research itself?

Yes?

MS. SONNENFELD: The hearing impaired -- the people with hearing losses, whom you have all over the place on the chart, were they wearing their hearing aids when they were tested?

They were not?

They were wearing earphones?

MS. ERICKSON: And they just...

No, they were listening with TDH 49 earphones, just the kind you use when you get your hearing tested. You just adjusted the volume control, the amplifier that we had, until it was comfortable for them.

MS. SONNENFELD: So they were at the correct volume for them?

MS. ERICKSON: Yes.

There was no frequency shaping, other than the shaping that was provided by the earphone itself. That slide I showed you goes out to 8,000-hertz fairly flat. They weren't listening with hearing aids. Listening with just ears covered, at a comfortable level, and not adjusted independently for the ears. Same at each ear.

MS. SONNENFELD: You did say that. I got lost after all those charts.

MS. ERICKSON: That's okay.

MS. SORKIN: I assume you used all adult listeners.

MS. ERICKSON: Yes, we did.

MS. SORKIN: There is one issue associated with that, that I hope doesn't throw too much of a monkey wrench into this.

Kids, in general, need a greater signal-to-noise ratio than adults. And the Access Board is, right now, also considering implementing acoustical standards in classrooms.

And in doing that we're looking at issues like what the signal-to-noise ratio should be for a child at the ear, what the reverberation time should be, et cetera.

So when we decide to adopt a standard for assistive listening devices, and I hope we do, we of course, would want it to be a standard that works for adults or children.

So, I just -- you might just keep that in mind.

The signal-to-noise ratio that's typically advised for a child with a hearing loss is 15 dB.

MR. BAKKE: That's an excellent point. And it reminds me -- where's that last slide?

The recommended standards that you have seen here are for discussion only.

We haven't made a decision that these are the standards we're going to recommend. These are for discussion.

We have put forward these standards to this body today, to talk about the issue, and we may need to do some more work on them. So that is a very good point.

MS. ERICKSON: It's an excellent point.

AUDIENCE MEMBER: I just have a question to clarify to make sure I'm understanding this correctly. Since the HVAC ambient noise in the conference room was such a factor, could it be that that as a variable in the classroom situation would make a difference from one classroom to the next as well?

MR. BAKKE: Absolutely. One of the huge issues in classroom acoustics is the HVAC and it's something that is often overlooked. It's a very, very important issue.

MR. GLICK: Ira Glick. Astute question, Matt: Signal-source speaker, correct? One source?

Why did you select an omni microphone rather than a cardoid (sp?) which, of course, would have cut down on ambient noise and allowed more distance and higher values?

There must have been a good reason.

MR. BAKKE: The reason we chose the omni is because it is the simplest, in terms of understanding what's going on.

It is certainly possible to use -- and preferable, in fact, if you're in a performance -- to use a directional or cardoid microphone.

The one way with dealing with the difference between those two is to look at what's called a distance factor for that microphone and you can take that into account.

So we felt that by keeping -- we had to choose one mike. We didn't want to add more variables to the situation. By using a good quality, top quality omni mike, we felt that was the best way to look at the data.

You can take the distances, actually, with a grain of salt, in that sense. You need to adjust them for the kind of mike that you're using.

However, what we're really interested in looking at is whether the Speech Transmission Index, as a general measure, would give you the best predictive value, and that turned out to be the case.

No matter what kind of microphone you're using, the Speech Transmission Index is very robust, in that it gives you -- it takes into account both distance or reverberation, room reverberation and background noise.

So regardless of what microphone you're using, the STI will be a good predictor.

MS. ERICKSON: You assume the STI would be better for a higher quality, directional microphone.

MR. BAKKE: Absolutely. If you use an omni, versus a highly directional microphone, the STI will be better with the directional microphone, because you've reduced the effects of the background noise and reverberation.

MR. JOHNSTONE: Being one of the people that has to make this work in the trenches, is it usually the case that when -- is it usually the case that when standards are adopted by the Access Board, they are typically for new construction, or would this also affect renovation beyond a certain level of -- a lot of times, it's the assessed value of the property; is that typical, for the Access Board?

MS. SORKIN: The Board's standards apply --

How's this?

-- the Board's standards apply to both new construction and renovation. And sometimes there is a different standard for renovation than for new construction.

But I can't imagine.

I'm speaking out of turn, but for this kind of situation, I can't imagine that we would have a different standard.

The other thing to remember is that: Communication accessibility actually comes under the Department of Justice. So we develop rules in the Access Board that becomes part of the guidelines, but it's up to the Department of Justice also in terms of how they want to administer the communication accessibility.

But it would be both.

And it should be, don't you think?

MR. ROSS: I'd like to ask the people from industry whether they think, you think .75 Speech Transmission Index is an onerous requirement for you?

It does seem to miss some of the people below that horizontal -- vertical line, oblique line. And I'm wondering if .8 would be something that would also meet the idea for children, as well as being a relatively easily obtainable by industry.

So, could you comment on that?

MR. FAST: Hi, Larry Fast from Sound Choice, assistive listening devices. I think this sort of throws it back into your court a little bit. Because most of the quality we find -- what we've seen in our products and almost everybody else that's out there, almost all the other companies, is that when proper installation and proper procedures are used, the proper microphones are attached, the system is not asked to do things it was never designed to do: Actually, they're all quite good.

I don't see that as being onerous, except that it comes down to the installer, the sound contractor, the facilities manager, to make sure they're doing with the system what it was intended to do when it does leave the factory.

MS. SORKIN: I think that's a really good point.

But what the board is, in fact, trying to do, is ensure that it's up to the facility manager to make sure that their system is operating in compliance with the law.

Right now ADA doesn't address how the system operates at all.

So if you, as manufacturers, are then producing a system that's designed to operate a certain way. And then the installers do something wild that makes it operate in a different fashion, it then puts the burden on the installer or the facility manager to ensure that it's operating the way that it should.

So I think, I would go with what's feasible, and then we can put pressure on the installer, and the facilities to ensure that it's operating the way that it should be operating.

Right?

MR. FAST: Yes.

MR. BAKKE: Before you comment, if you could just say your name, that would help our recorder.

MS. GLICK: Paula Brown Glick. I recently had some correspondence with the Access Board in which I complained that the courtrooms in New York City are not accessible.

I was finally informed by the director of compliance and whatever it is, that installation is all that they will be concerned about and not the actual effectiveness or facility's organization.

As long as something is installed, they have nothing more to say about it, is what they told me.

MS. SORKIN: Right now, that's correct. That's the way it works right now. All we do is say there has to be X number of receivers in the facility.

But what we hope will happen as a result of this research is, we will then be able to implement standards under the ADA. That is why Lexington is doing this work.

DR. LEVITT: I'm not going to say much, because I have a cold.

I think the STI is a very valuable index. It is a danger that manufacturers will focus on just the electrical components of an ALD system, that is from the microphone to the transducer, but the biggest interference comes from ambient noise and the reverberation in an actual room.

Now, if you work with an index like the STI you're going to need an instrument that measures it when you install something, now that might be onerous for the manufacturers, but I think has' the only way to go. And I think the answer will also be something like: You'll meet standards for people within a certain distance of a loudspeaker, say in a courtroom, or if a speaker is within a certain distance of a microphone.

And those have to be specified, because you can have a perfect system that won't work well if the speaker is too far from the mike or if the listener is too far from the loudspeaker.

MR. JOHNSTONE: Chaz Johnstone.

Glad you brought that up, Harry, that in addition to having an instrument. I'm curious as to who would then have to know how to use this instrument?

Is that something that facilities are going to have to either have on hand or contract to come in?

Will installers have to have that?

And then I think this afternoon we're going to talk about: How do you enforce behavior modification of the presenters?

And who is responsible for that?

MR. FAST: If I could just amplify on my earlier comments in regard to what's just been said: In the real world of the manufacturers, in many, many cases it is the third-party sound contractors and installer, and other people that are involved in construction or renovation that are doing the work.

I'm not familiar with how all the standards for ADA mobility installations go, but I know there are probably some fairly strict guidelines on ramping, door widths, many parts of which a contractor himself is responsible for, so for instance, the manufacturer of guard rails or other ADA compliance pieces of hardware, which is kind of what we are on the audio side, then sells to a contractor, that contractor in the facilities has to -- somebody is responsible for making sure that that was done right. I think it would be the same kind of sets of standards to be provided to the contractors and put them and the facilities people in the spotlight -- in any event --

(Announcement over the school's PA.)

MR. FAST: There's a typical situation, that if the contractor or installer takes what is a very well designed, compliant piece of equipment, puts the microphones in front of the air handler blowers, or in front of the PA speakers, somebody other than the manufacturer has to take responsibility for that.

MR. BAKKE: By the way, that's a horrible, horrible PA system. It is 30 years old and we are -- we've recently received funds through the Dormitory Authority. We're putting in a visual PA system that will have text, also sign language and quality sound. So that's...

That does not speak well of us, I'm sorry.

MS. STIERNBERG: There's another link in the chain that I think we should mention. Larry's comment reminded me of the Earler by Becket (SP?) lawsuit, where an architectural firm itself, was held responsible for the design.

In sound systems, very often, particularly in large projects, there are specifiers and acoustical consultants, who are actually drawing up the design and the plans, that the installers, then carry out. I think that's another link in the whole chain to consider.

MR. BAKKE: I feel like Oprah.

Andy.

MR. MUSEI: One issue, most of these projects are purchased under a low-bid circumstance. On that basis you don't always get the best installer or the best possible installation.

MR. BAKKE: Okay.

I would...

If that discussion has sort of petered out, what about the 12 dB?

When we looked at our data, it seemed to come out at 12 dB signal-to-noise ratio seemed to be appropriate, in that it is -- if you look at this set of data: Anything at 12 dB really stinks. So the 12 dB was offered as a lowest sort of common denominator. Maybe 15 dB, by adding a 3 dB sort of over there. We don't have a point at 15.

But even with an 18 dB level, we found the normal listeners were not too satisfied.

It's an interesting piece of data, actually, showing people with hearing loss less sensitive to the interference.

I can't explain that.

MS. ERICKSON: People without hearing loss were pickier throughout.

MR. BAKKE: We had picky hearing people, I guess.

It may be that it's not a psychoacoustical phenomenon. It may be that it's a phenomenon of what people are demanding. People with normal hearing may be much more demanding in terms of what they expect from a performance.

However, the point being 12 dB may be too conservative and we may need to talk about raising that.

I'd just like any comments people have about that.

MS. SORKIN: I would really urge you, in your report, to highlight that these were adult listeners.

MS. ERICKSON: That would really address the children issue.

MS. SORKIN: Just say that. When the Board makes decisions about what kind of standards it's going to adopt, it's going to look at your data. We're going to argue and go back and forth, et cetera.

I would really urge you to say that these were adult listeners. We know that children need a higher signal-to-noise ratio than adults and a lot of assistive listening devices are used by children in classrooms, obviously.

And so, although -- you know, that was primarily a study that was commissioned, because we were interested in ensuring that public places had appropriate quality levels in assistive devices.

These same systems are going to be used by kids in classrooms, et cetera, so I would really urge you to make that point and think about that in your recommendation.

DR. LEVITT: I think another important point is these were based on the average, which means half the listeners were dissatisfied. I think you should look more at the 90th percentile, find that signal-to-noise ratio in which 90 percent of the people are satisfied, maybe 99 percent.

MR. DAVIS: Bruce Davis of Gentner Communications.

I believe we have to make a clarification. I believe you're talking about the dB level in the environment, not for the products themselves; is that correct?

I think that distinction should be made in the report itself, so we don't get confused where we're trying to make a distinction with the product, and more with the environment itself.

MR. BAKKE: That's a good point.

The idea here is to look at the output, really the signal at the ear of the listener.

So even in doing the Speech Transmission Index, what you're interested in looking at is what's coming out of the earbud that goes into the earpiece or what's at the ear of the listener.

Thank you, that was very helpful.

I think the same issue that Harry just pointed out with regard to signal-to-noise, can be applied to Mark's question before about this Speech Transmission Index: We need to look carefully at what -- at the level at which, say 90 percent or so, maybe 95 percent of the people were satisfied.

Peak clipping is probably not too big of an issue. I doubt there are any systems that peak clip to the level of 18 dB.

The other question I had, though, was with regard to output level.

This is a little tricky. We've said 110 dB SPL because that appeared to be the level at which the most severely hard of hearing person wanted the speech to be presented.

The issue then becomes one of spillover. If I'm wearing headphones at 110 dB, the people next to me are going to be very upset.

So we need to think about this, not only in terms of what's acceptable for the person who is wearing the headphones, but what type of headphones can be used so that the spillover doesn't go to the next person.

Then there's the issue of using a neckloop or some other method of getting the signal into the ear.

So this is a little tricky.

Anyone have any comments?

MR. JOHNSTONE: Just one quick question: This is 110 at the ear?

MR. BAKKE: (Nodding head.)

MR. JOHNSTONE: Okay.

I guess the concern that I've received from people that I've queried, is that that might be a little too low.

But realizing the difference, just going up one dB, what about looking at something like 112 or -- I'd like to see 115, but I know that's tough to achieve when you get that high.

MS. GLICK: I use all kinds of assistive listening very frequently and I can give you my recent report: Last night we went to a movie and Ira complained that what I had on was too loud for him, when I was sitting there.

I also find that those that fit over the ear have more spill-out than the ones that fit inside the ear.

The infrareds that go into the seem to have less of that spill-out.

MR. BAKKE: Okay.

Very good.

We promised you, you know, our agenda, at 11:15 we would take a break.

It is now 11:15.

Thank you. There's coffee. The bathrooms are out in the hall, on the right side of the hall, toward the right you will find a ladies room and a men's room.

(Recess held.)

MR. BAKKE: Can we get started?

We've given you a five-minute break, five minutes over the 15.

Sometimes the conversation between the sessions is very valuable, but we do have to keep on some kind of time line or else we're going to get lost.

So let me introduce Mark Ross, again.

Mark is going to lead the next discussion.

MR. ROSS: Okay.

Can you all hear me now?

Good.

Before I go on to the next discussion on our agenda, are there any other comments on the performance standards?

We had two important recommendations to the Access Board. One was a min number of signal-to-noise ratio of plus 12.

And the feeling we have, as we listen to you, that this was too lacks. That we could go to, perhaps plus 15 or plus 18 and define this would encompass more people. In going along with what Harry said, should have the cutoff at percent the 90 percent percentile, rather than 50 percent.

The second point was that the STI of .75, which -- the data that we had kind of put up for discussion, that this was, perhaps, too lenient, and that we could go for a little higher standard than this.

Does that seem to be the consensus here?

MS. HO: Yvonne Ho. There is just something I want to throw out on the table. I was having a short discussion with Donna here, too, in consideration of all these standards, I think they are established as the minimum.

Of course, from a manufacturer's standpoint, it would behoove each and everyone of us to exceed these standards, ensure through the installation and actual applications that these standards are met.

Particularly in regard to the 110 dB SPL for the system, I think there's more qualification that might make this more reasonable, because when you look at the general population of hearing impaired, you're talking about 28 million people. But really in essence, between 6, to 7 million people who are actually wearing hearing aids. My guess is the number of people identified with hearing loss are of mild to moderate.

And this level at 110 would be too excessive for that population.

MR. ROSS: So you're thinking that 110 as a maximum, but would a volume control take care of that?

You know --

MS. HO: Yes.

MR. ROSS: Okay.

MR. GLICK: Ira Glick.

I think you're absolutely right, Mark.

I would -- slight contra distinction to Yvonne, say that the max volume should be sought, because if you are profoundly hard of hearing, as Paula is, it's impossible to really hear unless you're up very high on the dB scale.

And people who don't require that, are able to turn down to the lower scale with a volume control. And all ALDS do have volume controls.

MR. ROSS: I should --

MS. SORKIN: I was just going to say: That is really loud, though.

The conversation that Yvonne and I were having -- I said your name wrong, Yvonne -- that if you listen at that level for an hour or two and you have a moderate loss, and you do that on a regular basis, you're damaging your hearing.

I'm a little concerned about it.

At the same time, I understand that there are people that have that level of loss, but aren't they wearing hearing aids?

And they're using the device, hopefully, in concert with their hearing aid, not sticking it in their ears without the hearing aid.

Remember, there is also no correction from the device or what someone's hearing loss is doing.

I just have that concern, as well.

MS. GLICK: I do need to point out that not everyone has a comfortable level with hearing aids and I have never been able to use a neckloop, no matter how many times I've tried with no matter how many hearing aids I've used. It just isn't universal.

MR. DAVIS: Bruce Davis with Gentner Communication.

I'm not hearing impaired, per se, but I would imagine louder is not always better. I would think the quality of audio is more important than the loudness.

Of course, you have to have some certain level of loudness and quality in order to hear at an impaired level. But I think a modulation in the processing, within these products is probably more important than the actual very, very high level of sound that comes out of them.

I do believe we need to have some level that says, we need to at least achieve this level, but certainly the louder and louder and louder you go, it is harder to have that modulation and the processing correct with that level.

And a lot of people that we find that might complain about our products, are just turning that on, all the way up to the highest degree, where they'd be better offered backing that down to the 75 percent level. They just think the higher the better, I think we ought to be careful with how high, we say high needs to be.

MR. MUSEI: Andy Musei.

I suspect 110 dB is damaging, at any elongated time could be very damaging.

I agree that the 110 dB is really a high reference point.

MR. ROSS: I have to say I'm thinking of the nine blind men and the elephant. From an audiological perspective, the output of my hearing aid is 135, and I like more sound.

And audiologists who put a hearing aid on with 110 output, you're dealing with a mild to moderate loss.

But, thank you. Because we are dealing with different populations and it is important to remember that.

MS. ERICKSON: Mark, just getting back to the listeners in this study.

MR. BAKKE: Mike.

MS. ERICKSON: Listeners in this study listened at a wide range of listening levels, the recommendation for 110 being a maximum, people could still turn it down, but I think we should have it be a maximum and still maintain the quality, when you start getting above 110, you now have issues of quality and potential damage to people's hearing. These are for general population use, you don't really want to go much higher than 110, because it is potentially damages, hopefully if you had a lesser degree of hearing loss, you would wear it at a lower volume control setting.

I think 110 is -- certainly there should be a dynamic range over which it is able to amplify -- from I guess 70 to 110.

It should have a very nice dynamic range to accommodate a wide variety of listeners.

MS. SORKIN: Can I ask a question?

MS. ERICKSON: Yes.

MS. SORKIN: I would just like to ask the manufacturers who are here, what the maximum is right now in the systems that you produce.

MR. FAST: To be honest, it's really -- we don't manufacturer receivers. And it really is a function of the receiver, if the transmitter has been designed properly. You can only modulate the signal, whether it's FM or infrared, so far before it's going to break outside of the FCC bands in the FM world or simply overmodulate and become an unusual signal in infrared.

So the transmitter designs try to get the most signal on to that carrier. At that point, when it hits the receiver, it's the function of the receiver of what it does receive. What it does with that little tiny signal that it picks up in whatever the transmission mode, demodulates it: Does it have a hefty enough amplifier, are the ear transducers more conventional Walkman headphone type, ones that are really optimized to work in the ear canal?

Phyllis was just going through the specs that we had from some of the receivers, just taking a look, some of them seem to be around 116 already, from what they're telling us.

I don't want to dump it of on the other manufacturers, but we can only do as good as the receivers are.

MR. ROSS: I think we've heard everybody on this. I think we really have heard everybody.

Unless somebody has an additional point, we're still going to have to formulate recommendations, but based on this discussion.

MR. BAKKE: I just wanted to point out a little discussion that we had on the side here, regarding a different issue. It's regarding the use of the Speech Transmission Index to set a minimum standard for the quality of the signal.

One of the things that we observed in the research was that the normal hearing subjects, 10 of them, were actually slightly -- well, the same or slightly more sensitive to degradations in quality, than the people with hearing loss.

What this sort of points out, from my perspective, is that -- I think what's done typically with the sound engineer or sound contracting, is that there's an individual who listens to the signal in the venue that's being installed and makes judgments based upon listening. The data suggests that's not such a bad thing to do.

It may be -- I'm not suggesting we don't need a standard, I think the STI provides us with a concrete standard, so that if there are questions or complaints, we can go into that location and make a measurement that can back-up that this thing has met the standard.

But at the same time, a well-qualified installer ought to be able to know, just by listening, whether that standard is being exceeded or not.

So that's something that makes it, perhaps, a little less onerous in terms of having to do an STI measure at every location.

If you have a well-trained installer, they're going to pretty much know that they've exceeded that standard.

MR. ROSS: That's a big "if," which we'll talk about in a few moments.

MR. BAKKE: Right.

MR. ROSS: What I'd like to do now is kind of go on to the next topic for discussion. Let me ask everybody and remind everybody: When you make a comment, please give your name.

We do have a permanent record here. Everything that's being said is going to be reviewed by all of us. And so we do want to know who says what.

The next topic is, I'd like to kind of briefly discuss: We'll ask you for any questions of fact and corrections dealing with the state-of-the-art paper. Not dealing with recommendations right now, just the background information on the state-of-the-art paper.

I did receive comments from Ira, and I do appreciate Ira's comments.

I received comments from Drake, which I will incorporate. And I received comments from Oval Window, which is not here, gave me a great deal of information on loop systems. Just now I got another correction from Sound Choice, and thank you.

Is it Sound Choice or Sound Associates --

MS. WALD: No, Sound Choice.

MR. ROSS: Does anybody have any corrections or additional information that will deal with the factual aspects of the state-of-the-art paper.

Ira?

MR. GLICK: Ira Glick.

It was really with great pleasure that I saw your first draft, and I took the liberty of making 100 copies, and passed them out to people I think will benefit. I think it is probably the best thing that has ever been written on the background of the use and types of assistive listening devices.

MR. ROSS: I'm waiting for the "but."

MR. GLICK: No "but."

I think particularly on page 2, sentence "Assistive listening systems help by bypassing the acoustical space between the sound source and the listener." I get any queries about why is an assistive listening system better or why do some people who wear hearing aids find they get belter reception with an ALD, and this is a wonderful way to describe it. And I thank you for that.

MR. ROSS: Thank you. Thank you.

MR. GLICK: Be patient while I find the others.

On page 19 you mention the emitter is composed of a number of light emitting diodes. The more diodes, the more powerful the system. I understand there's work, and I know that Sound Choice uses the large sized diode, there is some work going on with products that are going to be launched with the next year, even with maximum interference in a room, one diode on an emitter is going to be able to cover a space of some 3,000 square feet, minimal, even up to 5,000. So I think there's more stuff coming in on that. Maybe single diodes, or maybe double diodes.

I think it should be imperative that the Access Board says that any kind of connection between the receiver and the person's ear be equipped so that a loop jack can be inserted.

I'm sorry.

So that a loop jack can be inserted.

I'm getting feedback. I'm getting feedback from the microphone. I'll hold it over here to minimize it.

The over-the-head headsets with the receiver on top, when they first came out from the company in California, did have a jack. But the current ones do not have jack, and I think the Access Board should require that any connection be equipped with jack sockets.

MR. ROSS: Ira, I'm going to discuss that on the coupling phase.

Just as a question of the background --

MS. ERICKSON: You have to get your microphone back.

MR. ROSS: Let me have your microphone.

At this point -- can you hear me now?

At this point, really basically, the background, Ira, we're going to go into, if you'll look at the agenda, coupling, and that's going to be an important part.

MR. GLICK: Page 29, third line from the bottom, I think where you say "attain with an FM system" really meant attain with a PA system.

MR. ROSS: I have that one, thank you.

MR. GLICK: Okay.

You don't want comments on earbuds or batteries?

MR. ROSS: Not yet, no. We're going to get into that, and we're going to get into that very soon.

Here, why don't you take it.

MR. JOHNSTONE: Just real quick, Charlie Johnstone.

Mark, my comments are part of the E-mail that you hadn't opened yet, and I can give you another copy today. So I won't get into all that.

I was just curious as to where or if soundfield systems should fit into this?

I know that's a whole other area. Where do people feel we are on that?

MR. ROSS: To answer your question: I don't know.

Soundfield systems are simply PA systems with a wireless feed. So I'm not sure where -- they're used now, extensively in classroom. But they're not used, as far as I know, in large venue, except as a PA system.

MS. SORKIN: Just remember the reason this study was commissioned, was that it was to give the Access Board some more help on how to more effectively develop guidelines for assistive listening devices, and right now the ADA, we don't talk about sound field systems. We just talk about assistive listening devices.

If people thought we should address that, then you can always petition the Board to look at some other issue.

But this is focused on the assistive listening device, where you pick up the source and it goes right into your ear or into your hearing technology.

MR. ROSS: I'd like to go on to our recommendations, which -- is that what you wanted to --

AUDIENCE MEMBER: No, I have one more comment, which needs to be made.

MR. DAVIS: This is Bruce Davis with Gentner Communications.

Two thirds of the way down, last sentence: ALD systems in really large venues, such as football stadiums, can also employ one of the commercial FM frequencies, in which case a license must be obtained from the FCC." That's not necessarily true. As long as they're within the power range of the original manufactured product, there's no license required.

If they exceed that or do.

MR. MUSEI: Ten a array, or something like that, in excess of the manufacturer's product, then there needs to be a unless, but not unless...

MR. ROSS: Good. Exactly the kind of correction I was looking for. That's not something I feel comfortable writing about.

If there's nothing else, I'd like to go on and discuss our recommendations -- other recommendations.

If you'll look at your sheet "proposed recommendations to the U.S. Access Board," on universal accessibility.

Now, we have -- remember what we're doing: We're not making standards.

We are formulating recommendations to go to the Access Board. And this is the basis for discussion.

So we're not coming at you and saying: This is what we should do.

We'd like to talk about it with you and get input from you.

We have orientation and when we listened to the consumers, and the consumers feel very strongly about this, talking about infrared, they would like universal accessibility with the same receiver in every venue.

All right?

So -- and this came up with the manufacturers as well, where at least in terms of FM, it was recommended that there be some standardized frequencies.

And there has been a defacto, standardized infrared frequent circumstances 95-kilohertz, for many years, which is now changing. At least, if it's not changing, it's now being added to.

So our recommendation is that every large-area venue provide a 95-kilohertz signal -- as a subcarrier on the infrared.

MR. BAKKE: Megahertz.

MR. ROSS: No, kilohertz.

MR. BAKKE: We put megahertz.

MR. ROSS: How come nobody picked it up?

You're not reading it?

It's 95-kilohertz.

As a basis for discussion and coming out of consumer input, we'd like to recommend that every large-scale venue provide a 95-kilohertz signal in their infrared. That is, that it be accessible, that it be compatible with a receiver which would receive a 95-kilohertz signal.

The objects to it had been that the new ballasts provide noise which made it necessary to provide a higher frequency carrier.

We also heard from other people, Audex in particular, who said that there are ways to filter out that noise with a receiver that would make that objection no longer relevant.

Discussion?

MS. HO: I guess before I comment, I first want to understand the request for having the compatibility of one receiver working at the 95.

Is that because there's a lot of personal receivers, personal infrared receivers that are used? Is that basically it?

MR. ROSS: (Nodding head.)

MR. BAKKE: Ira, do you want to comment?

MR. GLICK: Just what I was going to point out, a large number of those receivers, in terms of their own income, spent a considerable of money to purchase their receiver on 95 K. Also, really haven't been that many high-powered interference, that seems to have been lessened, and you wouldn't find those in motion picture theaters, anyway, conference meetings, things like that, courtroom, but large-scale use of 95 in motion picture theaters, is probably going to continue, although I know Phonic Ear is working on different frequencies, so are others. But I think it is imperative that that 95 be kept to service those people who invested in that receiver in that category.

MS. HO: I'd like to make a comment. Phonic Ear and other manufacturers, too, are producing up to 95, that's not the issue so much. But I think also down the road, we see a lot of trends in consumer electronic, where a lot of these consumer electronics, are producing 2.3, 2.8 megahertz, I think that's another reason why we also have been looking at other frequencies. Because as more consumer electronics are being used in the home, then we'll see more compatibility at that level, too.

MS. SORKIN: You might. Some of these issues, I think are FCC issues, and some are Access Board issues, you might just be really careful in your recommendations to distinguish between what's FCC and what's Access Board.

It doesn't mean you can't address something that's an FCC issue and recommend to the Board that they work with the FCC on a particular topic.

But just be careful to delineate who has jurisdiction over what.

In this case I think what you're saying is: You're recommending that the Access Board set aside -- I don't know that we can do that.

Just be careful when you make that recommendation, that it's reasonable in terms of what the board can do and what they can recommend to other federal agencies.

Eric mayer, Sennheiser.

The 95 has nothing to do with the FCC, it's just to modulate the frequency, that's why these ballasts, that have just walked in and stepped all over -- excuse me -- stepped all over the 95 K much.

We really had nothing -- that's why they could do it, was because there is no jurisdiction over who's going to use that to modulate the infrared.

Sticking with 95 and trying to use a filter set, which is what probably the other -- if they change the lights again, you're just going to have another problem again. So moving it to another frequency where nobody is, was one of the only ways to get away from it. I mean, I think sticking with 95, you're just going to continue to have problems, regardless.

MR. GLICK: Nobody is saying sticking to 95 as the only frequency.

MR. BAKKE: Talk louder.

MR. GLICK: I'm sorry, protect that frequency and use everything else you want, but don't throw that one into the garbage can.

MR. FAST: Larry fast from Sound Choice.

Speaking just from our business in the trenches, actually. And we deal more, not with the entertainment industry, not movie, and theater, nearly as much as courtrooms and government institutions. We also find a couple of factors come into play. One, we usually get called into when somebody is retrofitting the room, means they're putting in new lights. Haven't shipped a new installation 95 K since early this year. Everything we've been doing is nearly at 250-kilohertz.

The other part of it is the dollars are always very tight. Very tough to pry the money out of any governmental institution or any line-item budget to put any system in. And to have them go for a dual transmission system, where they know that they're not really going to be able to use one in their facility could be very tough. It's something to be carefully considered in making the recommendations. I can certainly see the validity in the entertainment markets and in the people that have invested in their own -- I might throw one other item that came to me out of the blue, while listening to this discussion: Which would be the receiver manufacturers, we're not one, but there are others out there, with the infrared receiver, I suspect it shouldn't be too difficult to make a standards converter where a 250 K signal could be at a low power transmitted at a little clip-on device, to enable the older low 95 K receivers, to be used. It might protect the investment, might be something that would be a niche market for one of the manufacturers, or somebody new should get into that.

MR. GLICK: Just to hitch hike on that, I think the next year, after April, there will be a receiver on the market for infrared use, which will be able to scan, just infrared channels, just as the Drake receiver scans FM. So there won't be any problem another the receiver end, I think.

MR. ROSS: Okay.

MR. BAKKE: Can we summarize?

MR. ROSS: Yeah.

MR. BAKKE: Is it sort of the consensus of the group that we don't need to step on this, that maybe this is something the market will take care of?

Yes?

That's good to know.

MR. ROSS: Okay.

I think the same logic holds into the next recommendation, the FM, universal accessibility.

We had the same point of view that says: When people, consumers buy a receiver, that they should have it -- they should be able to use it in any facility that uses an FM transmission.

So no matter what frequency is being used within the band of 72 to 75, and now the upper band as well, 116 to 117 mega herds.

AUDIENCE MEMBER: 216.

MR. ROSS: 216 to 217 megahertz.

MS. GLICK: That may be okay for a lot of different facilities. But in the movies that have a multiplex, each screening room has a different frequency and there's some complication. I have sometimes listened to the soundtrack in a different room than the one I was sitting in.

MS. GINSBERG: That's what I was going to complain about, I was on a tour, where we all had FM systems on, divided into groups, unfortunately, whoever had the microphone, it was interfering with the other -- it didn't matter what distance we were.

I have been to parties, where people had the same FM system and we hear the conversation.

I would like to be able to switch to something to be able to move it to a different frequency.

MR. DAVIS: Bruce Davis with Gentner.

I believe again, that's an installation facility situation. Most of the manufacturers do allow for spacing, adequate spacing between channels and frequencies, so that these receivers don't pick up, bleed over or other venues, from other meetings going on or from the street or other situations.

So there again, I think it's an awareness with the installer. Also working with the system once it's installed in the facility, would help alleviate this type of situation.

MR. ROSS: Let me just tell you a point here that came up in the last meeting and that was in your paper: It was recommended earlier that the manufacturers proceed void a set of standardized frequencies with standardized --

(Announcement over the PA system)

MR. ROSS: There be a shut-of on that PA speaker, whenever we want to use it.

MR. BAKKE: Feedback.

MR. ROSS: Okay.

I have to read, to remember what I just said.

MR. DAVIS: Talking about standardized frequencies.

MR. ROSS: -- made by Ralph and other people at our last meeting, recommending that there be standardized frequencies used by the various companies. Even if you didn't use them all, but that the ones that were used, would be standardized, and that there would be a standard designator. So No. 1 also means the same frequency, no matter what company is providing the transmitter.

How do you feel about that?

MR. DAVIS: I think in theory, that's logical. However, I think that can prohibit creativity and design with products, designating channel 1 always to be a certain frequency and going on from there. Because a manufacturer may have in his product, presets from the factory, where you're able to select back and both on the fly. Other manufacturers may not. It may be a factory set-and-leave-alone type of device. So I think creativity may be a factor here. You might limit yourself by doing something like that.

I do think, however, frequencies in the median points could be set and there could be some standardization there.

MR. ROSS: I guess what I meant was: If you had 72.9 as a center frequency, that that would always be designated by the same symbol by every manufacturer.

This is what the recommendations were made: Not to affect creativity in manufacturers, but to say when you use this particular frequency, that frequency has a certain designator, which is accepted by the entire industry.

Is that -- I did miss your point there?

MR. DAVIS: No. I can see where you're going with that. However, I believe you would need to have a lower number of finite frequencies set for that, so that you don't inhibit creativity, in how the processing is done or how many channels are available and things like that.

In other words, I wouldn't dictate to the manufacturer, exactly how to manufacture the product, in that area.

MR. GLICK: I think you probably have opened an interesting can of worms. It's always been my belief that the reason various manufacturers use different frequencies, in terms of the three numbers that follow the decimal point, is so that you can't have interchange built of the equipment, can't use it with a Phonic Ear receiver, therefore they get all the business vertically.

I think it would be a wonderful thing if there was total compatibility and pick and choose from anybody's equipment, because the channels matched each other, but I don't know whether you're going to convince anybody to do that, I'm sure that's not something to be arrived at by dictate, but since I have the floor, to go back to the FM system, I have no corrects in what you say, but suctions as to the language, I think it should parallel the FM, i.e., universally set of acceptable channels, and then give the numbers, 72, 76.

"Permits universal access," what does that mean, you mean want a receiver like the Drake used, universal? I think the word "universal" may be confusing in that particular context.

MR. ROSS: Let me say before Yvonne says it, I'll say something about Phonic Ear: Thirty years ago, when FM systems were first made, Phonic Ear and Telex, other companies came out with it, there were standardized frequencies. We could take the green/red Phonic Ear module and equate it with the No. 4 Telex and equate it -- that's the kind of thing that I meant.

MS. HO: Again, my question would be, on the compatibility, I don't think there's so much of a need to make the equipment compatible between manufacturer to manufacturer, but is the request to have a set frequency compatible for a personal system?

Again, that's my question.

And is that something that the Access Board would have think authority ruling on, compatibility with personal end user, own personal system bringing into the theater or whatever venue.

MR. ROSS: I think what we were talking about here was that there be transmitted, a set of frequencies within any large area system, that a user, who bought his own FM system, could get access too whatever theater they went to.

MS. HO: Yeah, that's exactly my question: Does the Access Board have authority to do that?

MR. ROSS: I don't know.

MS. HO: I agree with Bruce's point, that would have a tendency to inhibit innovations, because we're developing product, bringing it into the industry, that is always going to be compatible with what is going to exist today. We're not looking ahead as to what's going to be there tomorrow.

MR. ROSS: Good point.

MR. DAVIS: Bruce Davis, with Gentner again.

Just one more comment, I don't believe, and I understand from your point of view, as a consumer, that the manufacturers, in any way want to inhibit any of the cross-frequency, or compatibility.

Gentner's goal is to be compatible with other manufacturers.

We have a congenial working relationship with other manufacturers. There's no animosity and we certainly wouldn't to be congenial with those type of products, so the consumer can use them in different venue, we understand that is very important with the consumer, just so happens, early on in design and in our development process, No. 1, we didn't have information with respect to the frequencies. We probably didn't go out of our way to try to gain that, per se, and realize the importance of it.

But as we go forward, we understand that it is very important and we'll work to make sure that is one of our criteria.

So that hasn't been, I don't think ever was, the reason for the noncompatibility. It just so happens, in certain frequencies that does occur.

I think in most frequency, with respect to the manufacturers, there are a majority of the frequencies that do coincide exactly one, with the other.

MR. ROSS: I'd like to hear from the consumers.

MR. GORDON: Joe Gordon. As a consumer, Mark, I don't think you can make a distinction between a facility and a personal system. When I go to a facility that has an FM transmitter, I bring my own FM personal receiver. So you're mixing a personal receiver with a facility's large-room transmitter.

I would use their receiver, but as Ira Glick said, I believe, most of them don't have a 3.5 plug that I can plug my loop in. So I bring my own receiver.

I would like to be able to turn a dial on my own personal receiver with a loop, to use their FM transmitter.

Until I can bring my loop, I have to bring my loop and the receiver, and the receiver is not compatible, most of the time.

MR. ROSS: I think what you just brought up, was the next topic on coupling arrangements, which one of the recommendations being made was to have a universal plug, a stereo, universal stereo plug which would accept your loop.

MR. GORDON: My comment on that last point about a universal plug: I think we all know Sangen (sp?), that big company on the West Coast that makes audio equipment. I purchased an AM/FM TV radio, the size of a pack of cigarettes, and I had to return it for exchange because the first batch that came out did not take a 3.5 plug.

So I think somewhere you should require that the plug always be 3.5.

MR. GLICK: Or since the alternate choice is a submini, if the manufacturer doesn't want to switch everything, socket, et cetera, to the regular --

I'm sorry, the nomenclature, what's the submini that's used in your particular plug, on the FM?

Yeah, it's 2.5, but what do you call it?

Well, whatever it is: They could supply an adapter to fit that, if they didn't want to regear their whole thing.

To recap what I'm saying: For equipment that's sub mini, instead of micro, mini, which isn't raised, the 2.5, the manufacturer could supply with each package, an adaptor fit that. I go to Radio Shack and buy adapters by the handful to switch from 3.5 to 2.5.

MS. GINSBERG: I hear what you said, but I originally ran into the problem where I was using a neckloop and my neckloop was one that came with the Comtek and that's one size. I tried to put it into something that was by Audex: It didn't fit!

It fit in, but it wasn't getting the right connection. I found that if I have my Audex neckloop, I can't put it into my come text.

I'm finding out --

The other thing, I just wanted to bring up with the coupling arrangement: Don't forget, please, when you're looking at this, I don't use a neckloop anymore. I can't stand it, I use direct audio input with a Y-cord, I have to make sure that's my connection, I have a lot of friends using the patch cord with the CI, so they're going to be using the same equipment.

MR. GENDEL: Josh Gendel. In addition to the size of the connector, you also have to keep in mind, apropos what Judy said: They're not all wired up the same way.

You may have a stereo connector.

For example, no longer with us, but Seimens was marketing an FM system, very inexpensive FM system, the way they wired up the connector, it was a stereo connecter, but they wired up with a ring, a tip, a ring, and a ground, and the tip and the ring were connected together and then you had the ground connection and the result of which: If you plugged a Sennheiser neckloop into it, which is wired between the tip and the ring, it works.

If you plugged a Williams neckloop, which was a mono plug, it would short circuit, you wouldn't get anything, which is exactly, I think what Judy is describing.

So I think you have to say something about how the wiring actually is accomplished.

MR. ROSS: Very good, yes. Thank you.

Judy, we stayed away a little bit from requiring standardized -- a direct audio input, because every hearing aid takes a different kind of cord.

So we recommend a standard 3.5 plug and then you get your own adapter to your own hearing aid.

MS. HO: Just one last comment: I wonder if it would be satisfactory in handling something like this, when you're talking about a very specific requirement, that the word "typical" be used in the wording?

So then, again, you're not stating that every product must have this, but just leave that up to the manufacturers and the marketplace to deal with that. And just indicate that the plug needs to be compatible with common audio devices or something of that nature. But use the word "typical" in place of that.

MR. GLICK: Ira Glick.

Should you specify that coupling sockets be universal, i.e., they can be used with either a mono or a stereo plug. I know when Phonic Ears, for production of mono , but production was switched, so they would take either a mono or stereo plug.

MR. ROSS: What I'm hearing here is asking inform a standardized plug and wiring, but that what you're recommending is that this be phrased in a way that it be optional?

I guess I hear other people saying they would rather see it -- doesn't seem like a very onerous requirement to me, I don't know, to make it a kind of a recommendation rather than a suggestion.

I would lien toward saying this is the -- I would lean toward saying this is the kind of thing that should be standardized. But these are the people that make the final determination.

MS. SORKIN: Just a comment on that.

When the Access Board was originally soliciting public comment for the revisions in the ADA.

The ADA is going to be revised. There will be revised rules out, probably early next year.

This issue of the standardized plug did come up quite a bit.

I can't comment on that, because the rules haven't come out. But that was an issue that came up from a number of individuals.

And I believe it was part of the recommendations that came from the advisory committee for the ADA, that was a couple years ago.

So it's a long-standing comment. It stems from frustration that people have about trying to use their own devices.

I have yet to use a Sennheiser system, because I don't have a connect or cable that if it was that device, I'd love to try it, because I hear people say it's a marvelous system. But I have a cochlear implant and I have to have a cable that fits between it. I just don't have one that fits with the Sennheiser device, when I go to the Kennedy Center, I can't use the system.

Personally, I think it's an important issue for that reason: Because some of us have to use systems in a particular way.

MR. ROSS: Actually, what we've been talking about is coupling arrangements, which is one of the other topics.

And this is a relatively easy, I think, at least for me, set of recommendations to formulate: That there be a grade of coupling options for people who go to a large area listening situation, that they can listen either with earphones, neckloops, silhouettes, or any other form of coupling option, that this be available to them. So this is the recommendation that we're making.

MR. GLICK: Would it be possible for that recommendation to bar the use of single earbuds, as distributed with receivers, or even if you could extend it with that, double earbuds, any kind of earbuds, stamp on it!

MR. ROSS: Well, in the paper itself we made a comment about -- it pictured, that depicted an assistive listening device with the Monday aural earbud, and that really raises my hackles, personally, but that's personally.

I think I would like to see -- we cannot mandate everybody wear binaural system, as much as we'd like to.

The kind of coupling arrangements we've been recommending would be variations of earphones, neckloops and silhouettes.

And I would certainly -- and we did in the paper -- recommend against using earbuds, as a -- monaural earbuds, as a kind of typical listening arrangement.

What's the percentage, what other kind of coupling arrangements should be made?

AUDIENCE MEMBER: I was going to comment on his comment, so...

MS. GINSBERG: I run into frustration in my SHHH chapter, where I have -- remember, I have people who are using an in the ear canal hearing aids, that don't have T-switches, they start complaining about being able to find an adapter for them.

Some of them have an in the ear with a little T-coil, want to be able to put the thing over their hearing