Has anyone used ADRO circuit from AH?

[shortspark]

I have severe to profound loss and the newer, open air type of aids are
not appropriate for me. I currently have AHs Freedom 16 high power BTE
and was thinking of upgrading to their new ADRO technology.

I would like to know the benefits of this, how it is different and
exactly what ADRO is suppose to do (beyond the advertisements). Does
anyone have experience with ADRO?

[e...@mindspring.com]

I too am SevereProfound and previously had the same aids from AH.
switched over to ADRO and am very pleased with them.

In a nutshell: ADRO does not use the conventional compression to
control loud sound(s). It works by controlling the output in 32
channels. The aid analyses the sound in each channel and changes
each channels output to match your particular Comfort Levels. So the
aid is running in a linear manner producing virtually no distortion.
Music is astounding.

Fitting is very different from conventional aids. Ed

On Fri, 5 Jan 2007 04:21:17 -0600, mar...@webtv.net (shortspark)
wrote:

[Ken]

ADRO seems to be a widely-used term - a search of Google throws up
several hits which suggest it is a common strategy in telephony and
hearing aids.

ADRO is one of the programs offered by Cochlear Corp for implants and,
I gather, one of the most popular. They say: 'This program will adjust
itself automatically when there is competing noise'

My impression is that it operates a bit like noise reduction headphones
- identifies repetitive background noise and subtracts it from the
sound delivered to the hearer. The above post suggests that it may do
this channel by channel - in the digital world this means applying
some mathematical process to the stream (or streams) of digital data.

I can say that the CI ADRO certainly achieves some background noise
reduction in motor vehicles and in city streets. But not much, but some.

[jay]

I had Freedom 16 aids as well, and upgraded them in September with the
ADRO technology. I was super pleased with the Freedom 16s, and hoped I
wasn't making a mistake. Well, I didn't. The new aids are amazing. I
usually leave them in the "normal" program, but do have a "restaurant"
and "music" program. I've found that I will switch into the
"restaurant" program in a very noisy place. Otherwise, the "normal"
program is perfect.

[Mason C]

On Fri, 5 Jan 2007 04:21:17 -0600, mar...@webtv.net (shortspark) wrote:

I have one AH BTE. I upgraded to the ADRO version.

My experience is that the ADRO is definitely superior.
The method of fitting, the software, and the result are advances.

It uses quite a different method of fitting. You can do it yourself without
reference to your audiometer chart. This is done by setting equal comfort
over a range of frequencies at a normal hearing level (not at the minimum
detectable as is customary). This comfort level is then the target
for the aid's adaptation -- done in frequency channels.

The new software gives detailed fitting instructions.
My impression is that the adaptation makes the fitting less critical.
I recommend the AH ADRO upgrade.

Mason C

[e...@mindspringl.com]

Ken: I don't think you have it quite right.

ADRO is the patented software (firmware) program from a highly
respected R &D firm from down under. It does not operate as a primary
noise reducer. However, the America Hears ADRO hearing aid does
include a separate noise reduction circuit using a large number of
noise channels.

ADRO constantly checks each of 32 channels and using the ADRO
algorithm determines the proper gain in each channel for maximum
comfort. This comfort level is based on the your response to test
signals during the initial fitting..

All the rest of the aids on the market use compression to set the
comfort level. Compression is ok in very small amounts, but for many
people that need lots of compression to control their recruitment, it
results in distorting the speech and butchering music. (Recruitment:
Uncomfortable response to loud sounds) Ed

.On Sat, 06 Jan 2007 02:17:57 GMT, Mason C

[Ken]

e...@mindspringl.com wrote:
> Ken: I don't think you have it quite right.
>

Thanks for your singularly lucid description of this technology.

[Medman]

e...@mindspringl.com wrote:
> Ken: I don't think you have it quite right.
>

Thanks Mason:
You are exactly right. However, there is nothing new in the ADRO
technology that other vendors do not also have i.e. having the DSP chip
monitoring channels without compression. This technology is available
on many vendors' hearing aids. The channel monitoring procedure is part
and parcel of the DSP function of the chip. The programming done in an
audiologist's office sets each channel to enhance or suppress specific
frequency responses based on the hearing loss curve of the wearer.

ADRO looks great as a PR tool, but it really is nothing new.

[jim]

Actually there are a few important differences that the ADRO has
brought along.

The DSP functions do consist of the common multiband equalizer as you
mentioned. And then for conventional DSP systems there is also the
so-called "compression". Of course if they really compressed the audio
waveform in real-time, hideous distortion would result. As a result
what is commonly called compression for hearing aids is really a form
of what radio technicians called AVC (automatic volume control) and had
been used to control volume for years.

In DSP based hearing aids each channel has its own "AVC" circuit that
operates on the slower "envelope" of the audio rather than on the audio
itself. All the implementations use a method of feed-forward gain
control (or volume control). This means that the channel's audio level
is detected in advance before amplification, then the "amplifier's"
gain is set to provide the desired gain for that level (in keeping with
the desired compression curve), then the audio is "amplified." This all
of course is done digitally in DSP fashion.

But for these conventional hearing aids, all kinds of other processes
(noise reduction, microphone channel gain/phase control, feedback
reduction, etc. all get to take a whack at those same gain (and phase)
controls. So more often than not the critical compression
characteristic the user wants matched to his recruitment characteristic
is sent way off into left field and the user's sound is compromised.

Apart from all the ADRO buzzwords Wedman is rightly suspicious of,
there is an important difference I tend to like. In these ADRO system
implementations (there's more than one for sure) they forego the
feed-forward gain control ("compression") approach in favor of a
feedback approach. In other words, they detect the OUTPUT level (not
the INPUT level), and based on that value, they set the channel's gain.
Using this approach, the effect of all the other shenanigans on level
are taken into account and the resulting "compression" curve can be set
to more constantly and more accurately match the user's recruitment
characteristic. I think that's a much better approach than the standard
feed-forward gain control approach and could be quite noticeable.

The ADRO systems mix in quite a bit of their own fancy dynamic gain
manipulations also but in the end the resulting level is forced to obey
the desired "compression" characteristic - that is it is forced to stay
within the user's dynamic range and (ideally) to consistently match out
his very non-linear loudness perception characteristic.

At least that's what I think it does. So it is significantly different.
Maybe it's better, maybe not. If ADRO becomes more widely adopted there
is bound to be many ADRO flavors. I don't have an AH ADRO instrument
but AH/Gennum seems to be doing pretty well with it so far.

AH also mixes in a new in-situ fitting process (Phonak Savia Art just
came out with one also) - and that I think is really that way to go.

[Ike]

Jim, thanks for this illumination -
I've been using Canta 7 BTE power, which I periodically 'fit' myself,
and might try the ADRO as a step forward.

Ike

[edz...@mindspring.com]

Jim: as I understand it there are other major differences between a
standard WDRC compression system hearing aid and the ADRO . These
have to do with the timing and algorithm to determine the applied AVC
(AGC). In standard WDRC compression they have a pre-set level for
the knee point to start the application of the control voltage and a
setting for the amount of the AVC (compression ratio). Typically this
is done in 4 to 8 channels. And the timing is selected typically from
a choice of three, that is, Syllabic (fast), "Regular"? (slower) or
a combination of both. ( I forget what they call the slower slew rate)

With the ADRO the AVC is applied in 16 to 64 channels (Typically 32)
for much greater precision and noise control. Then the amount of the
AVC is determined in each of the channels by an algorithm based on the
users pre-determined Most Comfortable listening Level. If the output
levels exceed a set level for a set percentage of time the Output is
reduced relatively slowly....and if the output level is lower, based
on the user's THL, in any of the channels for a percentage of the
time, the gain in that channel is kicked up slowly. Typical timing
applied is 3db/second. These algorithms are based on R&D by the
patent holder. They should handle most recruitment problems.

AH has some trick they use to get the through-put time (transfer rate)
down to 1ms. This is important in reducing low freq. distortion. (All
things being equal tranfer rate is proportional to the number of
channels when they FFT....more channels usually slows down the
transfer)

To handle loud fast transients the AH ADRO's have a superfast
clipper/limiter.

Granted there are similarities between the systems, but I think the
ADRO would deliver much cleaner sound the majority of the time in the
majority of real life ambiences. If other mfg's use ADRO I would
assume they will for promotion purposes use their own algorithms and
thereby make claims of superiority.

Jim, et al: I think I have this correct. Let me know if I got it
wrong. Ed

[jim]

Hi Ed

Great description of ADRO. Thanks!

My take on the terminology is that neither method is true compression
and both methods are complex, "smart" volume control approaches. Both
methods really stir in all kinds of factors (dynamic and static) with
regard to the need to fit a 100+ db dynamic sound sound into the user's
much smaller, "non-linear" dynamic range. Characteristics of the sound
trigger similar actions but trigger them differently with regard to
speed and extent. Both methods do (or should) take into account the
user's loudness perception curve and it's measured HTL, MCL and UCL
points. For sure, even if the the actions are structurally similar, the
dynamics are different and (as usual) the descriptions of the
philosophies bring along their own set of buzzwords.

I do see that the dynamics philosophies and response times are quite
different. That is, the "smart" part is quite different, particularly
with regard to time constants. As you say there are many more channels
for ADRO. They also indicate extremely low delay even given the number
of channels. Without more detail we are left to guess. How do you get
32 channels with only 1 ms delay?

My guess is that this harks back to the original application
(invention) of ADRO where the output went to 32+ CI transducers
(electrodes) instead of just one receiver. You were "forced" to have 32
channels! But here it may not have been necessary to have 32 "narrow"
channels. In other words they could be rather wide, shallow-skirted,
highly overlapping channels. In this way they would emphasize certain
sections of the spectrum collectively but not with the definition you
think of when you envision a 32 channel equalizer. So (to extend my
guessing) when they decided to transfer this technology to conventional
hearing aids, they just summed the 32 outputs (DSPwise) and drove the
receiver with it. The upshot is that the filtering is probably much
more gentle than what we think of when we think of a 32 channel EQ.

^ Pure guessing on my part.

So (right or wrong) I'm left with the impression that the feedback AVC
architecture I like so much is the major STRUCTURAL departure for ADRO
from conventional design and the dynamic control strategies are a
carefully developed optimization of this kind of a filtering system.

It's important to recognize that ADRO being the offspring of a long and
successful CI approach has a lot of development work behind it. This
isn't just another tweak and buzzword to make the marketing folks
happy. Like any departure though I would expect to hear pluses and
minuses from wearers of the new technology. So far things though look
pretty good.

[shortspark]

Thanks to all for the excellent responses. Although much of what has
been said is beyond me, I at least have a better idea now of what ADRO
is and a few good recommendations as well.

So, I called AHs and they said my hearing audiogram was two years old
and they would like to see a new one before saying the ADRO upgrade on
my Freedoms would work for me. I contacted my audi and had a new test.
Basically, nothing different than two years ago. I sent it to AHs and I
will wait to see what they say.

My audi is a pretty fair and I think honest fellow, and I've been with
him for some time now. He is the one who fitted me with the Siemans
Infinity a few years ago. I did not mention AHs but asked him about the
new Siemans model I've been reading about, Centra. He said it has been
out a short time and he has not sold a single set yet. He felt that the
number of channels might be a slight benefit but not a great one as most
of them would be shut off anyway for my hearing situation. He said he
really does not believe that, for me, there is any aid out there today
that will be such an improvement over my Infinity to the point that when
I use it I would go "wow". In other words, there would be very minor,
if any, noticeable improvement and the cost for this would be many times
greater and could go easily beyond the five or six thousand dollar
range.

The upgrade of my Freedoms to ADRO should run less than $1400 so maybe
that is they way I'll go if AHs thinks it will do any good. I will
report back here as to their recommendation.

MARTY

[ed...@mindspring.com]

Jim: Your educated "guesses" sound to me like they are right on
target. The skirts of the 32 channels sure could be broad, based on my
experiences in noisy restaurants, where the overall HA level drops
sharply. Fortunately, for all, in that kind of ambience, everyone
virtually shouts. I find the ADRO approach about equal to or slightly
better in noise then the standard high end aids I have. Nothing seems
to work better in noise than directional mikes with high DI.

As I mentioned before the ADRO approach seems to me to be vastly
superior with music. And the one on one in quiet is somewhat clearer
with the speech cues more intact.

I have recruitment up the gazoo. So, another difference is such
everyday stuff like running water in a sink. With the ADRO the
annoying sounds just melt to a comfortable level. With the
conventional aids it remains an annoying distorted clatter. Ed