Difficulty hearing the beginnings of words in soft speech - what to change?


#1

Hi all,

Currently trialing Quattro’s. I really want to get these to work. In challenging environments, these things are shining. At home and soft speech - not so much, and it may be a deal breaker. I love the streaming & connectivity of these Quattro’s, though, so I am fighting to get these things to work. I have been REM fit and this week will be my 4th adjustment.

To those who are familiar with SmartFit: What sort of impact could Binauaral directionality III w/ spatial sense have on this? What does “directional mix” do in regards to being high or low? I don’t have a NoahLink Wireless, but I have poked around in the fitting software to investigate what could be changed for my appt this week.

I could have the Audi adjust the noise tracker II to be by environment, but part of me thinks there’s more to it than adjusting the background noise aggressiveness.

A lot of times my wife is near me but not directly in front of me, and it’s in this situation where I can’t understand what she’s saying. For example, she said can, but all I heard was “an” and I had to use context (and a lot more effort) to communicate with her.

Sometimes I feel the Quattro’s are too aggressive in background noise. Our furnace is not that loud at all, but sometimes I find the Quattro’s have automatically put themselves in moderate or considerable noise reduction.

Any thoughts?


#2

Keep good notes and tell your audiologist what you’re telling us.


#3

The attack and relaxation of compression can affect relative loudness of soft and moderate tones in speech. The default setting for all speech programs is “Per Syllable,” but there are also Fast. Moderate, and Slow settings (music is Slow for other reasons). An optimal setting raises the loudness of soft tones while not affecting moderate ones much. You might ask your audi or have her ask a ReSound rep, how the Fast time constant compares to the Per Syllable and whether there is any further desirable way to influence soft speech thru this setting. I think the standard ReSound recommendation. though, is to have your wife wear a remote mic.

Am grocery shopping. Will add link to Starkey compression manual when I get home.

Edit_Update: This is the text that I am referring to about attack and release times for implementing compression to input sounds, from p.16 of the Starkey Compression Handbook-Fourth Edition, Copyright 2017

AT = attack time, RT = release time, CR = compression ratio

“For time-varying inputs, such as speech, the effective CR is significantly affected by the AT and RT. When
the AT and RT are fast – i.e., shorter than the duration of a phoneme or syllable – the gain changes sufficiently quickly to amplify softer components more than the louder components. The result is an effective CR for speech that is similar to that specified on the basis of steady signals. On the other hand, when the AT and RT are slow – i.e., longer than the duration of a typical word or utterance – the gain does not change much between softer and louder phonemes.”

But then your other problem is this (since you have a serious hearing loss where undoubtedly high compression ratios are implemented (later on same page):

“Another example of the interaction between compression parameters is the observation that fast ATs and RTs are more detrimental to the perceived sound quality at high CRs than at low CRs (Woods et al, 1999).”


#4

The other thing I’ve asked my wife is please don’t just start speaking with the important stuff but get my attention first, look at me, and maybe raise her voice a bit (in days before HA’s)


#5

I have found that by tweaking the frequencies (boosting the higher ones) my comprehension of speech is better. It’s those “f” “S” “th” kind of sounds that can get lost in the shuffle. But I also like jim_lewis’ suggestion of having the person get your attention or stand facing you! What are we? Psychic? We rely on VISUAL cues as well as straining our weary ears.


#6

There are gain settings for soft speech, medium speech, and loud speech. You can ask for soft speech to be increased without affecting medium or loud speech. I always have to have soft speech bumped up a couple of times.


#7

The default Speech Clarity settings for the Quickset in the All-Around (the all-purpose ReSound program) and in Outdoor boost midtones by about 5 dB and treble by about 6 dB and reduce base 2 dB relative to the regular settings for these phone app programs. My hearing loss is very different than mjstew33’s but for me, the way ReSound Smart Fit programmed me initially, the Outdoor program is closer to my prescribed fit than the All-Around program so the Speech Clarity quickset works better for me with that program. mjstew33 could twiddle with the noise reduction and bass, midtone, and treble settings of any basic program and save the altered settings as a FAVORITE (labeled “WIFE” (?) in the ReSound app). Since he’s a very clever M.D., perhaps he’s tried this already with unsatisfactory results?


#8

Here are the built-in time constants for attack and release of compression used in ReSound HA’s like the Quattros.

The figure is taken from the supplementary material provided for the Audiology Online course by ReSound:

ReSound Smart Fit: Advanced Features Tips and Tricks


#9

Edit_Update: I watched the course below and in the video presentation for Mr. Emm, the presenter does specifically discuss changing the gain for Soft Speech environments using the Per Environment feature of Environmental Optimizer II and at the same time, using NoiseTracker II, making Noise Reduction in soft speech environments more aggressive (you mentioned the refrigerator humming in the background at home). The actual video has a lot more details and points than shown in the slides from the presentation that I put in the post here). End_Edit_Update

The following slides are lifted from an Audiology Online course by ReSound entitled:
Grand Rounds with ReSound (Smart Fit Update): What to do About Noise?

The slides come from the course handouts. Presumably if you actually watched the course, you get more explanatory video and audio. “Mr. Emm” is one of three problem patients in the course that ReSound is coaching audi’s on improving results. The course is based on the LiNX 3D but presumably the approach for the Quattro’s would be very similar.
image

image

image

Hope this helps! Good Luck!


#10

Is this a success story? I am married since 20 years now but my wife still starts talking before getting my attention. Sometimes I don’t even realize that she has stopped telephoning and switched over to talk to me :frowning:
As a software developer I describe it as: She is communicating by using UDP instead of TCP


#11

Wives are notorious for speaking from 2 rooms away, then getting miffed that you can’t understand them. My wife pushed me 7 years ago to get hearing aids (thank god), but still starts talking with her hand in front of her face or from the kitchen or while my streamer is on etc. My friend just got his first hearing aids and is doing pretty good except for his wife. He misses the first few words she says.
Signia has own voice processing and they take your voice and lower the amplification. Why can’t they have most important voice processing and take someone’s voice and bump it up? It would solve a lot of my problems if I could hear her clearly.:thinking::thinking:


#12

You can download a free sound meter app and after several tries determine where her normal voice is, soft speech (around 50db), medium speech (around 65db), or loud speech (80 db), and ask your pro to bump up that range a little.


#13

The ReSound fitting advice for folks having trouble with soft voices advises folks to do exactly what you’re suggesting, Don, and what you have found to work well from your own experience. One point the software advice reminds folks of is that the gain change does not just crank up the volume of the soft voice that you want to listen to but it also cranks up the volume of all other sounds in the environment. So the ReSound advice also recommends that the pro crank up noise reduction a notch or two at the same time for soft speech environments to help improve signal-to-noise ratio.

Interesting that there is not much discussion on how noise reduction is achieved - maybe because all HA OEM’s leave out the important details and/or inflate the claims of how well their algorithms actually work. ReSound claims that their noise reduction algorithm looks at the sound in the gaps between speech, comes up with an average noise pattern and subtracts what it figures is the average background noise from the combined spectrum of speech and noise. Maybe all HA’s basically work this way and each OEM just adds on a few unique tweaks?


#14

So are husbands, frankly. Whichever party doesn’t have the hearing loss. Although I find to that the person WITH the hearing loss will still often call from two rooms away, which doesn’t help because then their partner will respond from two rooms away and they’ll still miss it.

I’m a bit surprised that in this cellular age, people aren’t just phoning eachother in those situations.

Yeah.


#15

The ambient AI assistant devices such as Echo and Google Home have intercom features. Some folks say that they don’t need or want the ability to interact with an AI assistant in their HA’s but with the ability to hear responses through your HA’s, such an intercom feature with a significant other a room or two away might be very useful and alleviate having to walk halfway across the house to find your spouse (chosen to rhyme!). :grin:


#16

LOL! Great laughs and suggestions on this thread. I like the idea of adjusting the low, mid and hi frequencies for optimum comprehension. Wish I was techie enough to get some kind of Marvel app and DO that. I rely on visits to my aud-guy …

Not sure why, but having heard the HE says SHE says arguments for so long from so many people, I’m convinced that there is some kind of “selective hearing” symptom between partners. It’s closely related to the other condition “convenient amnesia” as that pertains to chores and anniversaries. :smirk:


#17

My problem is the opposite. My wife’s speech trails off toward the end and she tends to sub-vocalize there, making it difficult for me to understand the last part of her speech. I doubt I can change her speech pattern after all these years. I ask her to repeat, usually just the last part.


#18

I didn’t mean to imply women are at fault. My son and nephew are impossible to understand. They are 30s and 40s and mumble. I’m sure wives have the same problem, my wife does when I answer her back from a room away, and she aces audiograms. BUT if I want to start a conversation with her, I stand right in front of her and speak plainly. She knows she is short with me when I can’t understand, but I do ignore her talking to herself or the tv. But sometimes I miss something important thinking she is just talking to tv. Getting someones attention before asking the important stuff is all it takes.:raised_hands:


#19

Maybe from a post by Don in another thread I thought that Don was also suggesting that the typical frequency range in which a spouse speaks for at least the most difficult to recognize phonemes be studied (as well as the dB volume of such utterances) and the gain for the typically soft volume of these utterances be specifically turned up at these frequencies/frequency ranges. Not sure if, for instance, with the ReSound software that one just turn up the gain for soft speech whether it just turns up the gain for specific frequencies at the “soft” input level or whether it by the board turns up gain for all frequencies across the board. Not sure, too, whether the effect would/should show up in the fitting software gain table if the gain is just adjusted on a PER ENVIRONMENT basis by hidden logic in the fitting software/HA firmware.

If anyone knows if per environment settings show up in general in fitting software gain tables, thanks for the info. That would be useful information to know, particularly if one could fine tune gain setting for particular environments that way. Maybe the gain settings table is only and always a PER PROGRAM table and the particular gains for an environment under that program are just behind the scenes (proprietary?) tweaks, not to be revealed.


#20

I’ve found that many people from MN and other areas with Norwegian influence raise their voices at the end of sentences (think of movie Fargo) and this is far easier for me to hear. Unfortunately, while my wife once lived in MN and can mimic their speech, she doesn’t ordinarily speak that way. In regular English lowering the voice is the signal for the end of a sentence.