Home theater for hearing impaired

[ataloss]

I have a 10 year old home theater set-up, with a decent B&W center channel. Though a bit muddy sounding, it has served us well. Now my wife's hearing is going, and she often cannot make out the dialogue. I am looking for a decent center channel, up to $1000 (but closer to $500 would be good) that produces especially crisp dialogue. I guess that means it would have a "bright" sound, but, if that's what it takes for my wife to be able to hear, then that would be fine. I have looked at the Klipsch Reference Series, B&W CM Centre, and Axiom VP150 and 180 for starters, but could use a lot of help, still. Coming from a small town, I'd like to have a good idea of what to focus on when I get to a decent home theater store.
Thanks

 

[STRONGBADF1]

I would guess that it depends on how her hearing is going. What frequencies are most affected. I don't know for sure...I'm not a doctor and I did not stay at a Holiday Inn Express last night.

I have a problem with background noise making things unintelligible. (don't play drums without hearing protection) Saying that, the clarity of the dialog on my SVS MCS-01 is very good.

Other options could be room treatments and or equalization. Turning up the center channel a little could also help. Last but not least would be headphones.

Hopefully someone with some experience with this will chime in.

 

[JerryLove]

Try diabling the center and running phantom. That did a lot of good for me and costs nothing to try. Another option is to get an eq and modify the sound going to the speaker. This is easiest to do either wit an eq built int the avr or if the avr has per-out and amp in.

 

[ahblaza]

Try diabling the center and running phantom. That did a lot of good for me and costs nothing to try. Another option is to get an eq and modify the sound going to the speaker. This is easiest to do either wit an eq built int the avr or if the avr has per-out and amp in.

Jerry, I agree with the phanton center, much improved over the cheap center I was using. Could you explain how to EQ the center within AVR built in EQ, which frq to play with? Thanks

 

[ataloss]

A little more info

Thanks Strongbadf1 and JerryLove,
A little more info -

It is the mid and lows that are especially dropping out of her hearing. I guess this is why I think bright might be better. It has not helped to turn up the center channel. Disabling the center channel and running through the rest of the surround sound system has worked better for my wife in terms of her being able to make out the dialogue. It is one of the thins that makes me think that a crisper sound might help. Recently, however, that solution is not cutting it either... and we lose the full surround experience, so I'd like to improve on that situation.

 

[moggi1964]

I second the headphone suggestion unless you both interact vocally during shows/movies.

It might take a bit of getting used to but the difference for your wife will be amazing I would imagine.

Hope you find a solution that works for you both.

 

[jostenmeat]

Thanks Strongbadf1 and JerryLove,
A little more info -

It is the mid and lows that are especially dropping out of her hearing. I guess this is why I think bright might be better. It has not helped to turn up the center channel. Disabling the center channel and running through the rest of the surround sound system has worked better for my wife in terms of her being able to make out the dialogue. It is one of the thins that makes me think that a crisper sound might help. Recently, however, that solution is not cutting it either... and we lose the full surround experience, so I'd like to improve on that situation.

If the mids and lows drop out, why do you think the highs need to be pronounced? Just curious about the logic.

Audition speakers using vocal stuff only. Like choral works, or whatever, singers, etc.

From the bit I've learned about speaker design, a horiz MTM will not get you the best chances of dialogue intelligibility. Vertically arrayed will have the best chances. You keep it vertical.

Then having the xover between tweeter and mid to be above what some here might call the critical dialogue passband. IOW, you don't want the complexity of a xover occurring right in the middle of the speech discrimination band.

Lastly, if one cannot make out dialogue, who really gives a rat's azz about surround effects. Focus on dialogue only, for now.

Are there are any chances that you can have a vertical center?

If not, my vote is heartily towards phantom, because you have not only your vertically arrayed speakers, but because the odds that the placements away from boundaries that may cause masking midbass boosts or strange reflections and/or diffractions may give you the best chances. That is not to mention the offaxis lobing of the aforementioned horiz MTM.

You might want to state a budget so that others have better recs for you. Perhaps size constraints too, just in case.

 

[JerryLove]

Jerry, I agree with the phanton center, much improved over the cheap center I was using. Could you explain how to EQ the center within AVR built in EQ, which frq to play with? Thanks

The mechanisms? That depends on the particular AVR and you'd need to check your manual.

As to what EQ to do? I'd start with a +6db boost around 1khz.

 

[JerryLove]

It is the mid and lows that are especially dropping out of her hearing. I guess this is why I think bright might be better. It has not helped to turn up the center channel. Disabling the center channel and running through the rest of the surround sound system has worked better for my wife in terms of her being able to make out the dialogue.

That second part tells me you are likely having the same sound issue I had. I would suggest continuing to run without a center (my center was easily on par with my mains and had the same issue).

I did not suggest turning the center up, especially given that it is now removed.

Dialog exists between 350Hz and 3500Hz. I'd look at increasing the relative volume of that range... I'd start with 1khz. See what improves clarity for you.

Before you do (if you have not) run your room correction (Audigy, etc).

It is one of the thins that makes me think that a crisper sound might help. Recently, however, that solution is not cutting it either... and we lose the full surround experience, so I'd like to improve on that situation.

I think it's a mistake to try to find speakers that are broken in a pleasing way. EQ the speakers with the actual EQ.

 

[jostenmeat]

Dialog exists between 350Hz and 3500Hz. I'd look at increasing the relative volume of that range... I'd start with 1khz. See what improves clarity for you.

Well, 3.5k hz is pretty high (think higher than the very highest fundamental that a violin can play, past the fingerboard on highest string, IIRC), but I understand that there are harmonics that make up our voices. That said, there are many men who can sing below 80hz fundamental, believe it or not.

I am not saying that 350-3500hz is not a more important range to look at (because I don't know enough about audio reproduction), but when fundamentals are concerned, it's way below 350hz.

 

[Pyrrho]

I have a 10 year old home theater set-up, with a decent B&W center channel. Though a bit muddy sounding, it has served us well. Now my wife's hearing is going, and she often cannot make out the dialogue. I am looking for a decent center channel, up to $1000 (but closer to $500 would be good) that produces especially crisp dialogue. I guess that means it would have a "bright" sound, but, if that's what it takes for my wife to be able to hear, then that would be fine. I have looked at the Klipsch Reference Series, B&W CM Centre, and Axiom VP150 and 180 for starters, but could use a lot of help, still. Coming from a small town, I'd like to have a good idea of what to focus on when I get to a decent home theater store.
Thanks

Hopefully, you have everything properly set up, with the levels properly balanced. If not, do that first, because if the center is set too low, that could be your entire problem with not being able to understand dialog. If necessary, you can also boost the center channel slightly (perhaps +1 to 3 dB) to emphasize it over the other channels, which, if done moderately, should not sound bad for you (though not ideal, obviously). You can also use an equalizer to boost the frequencies relevant to human speech in the center channel if necessary (though I would probably not do both that and boost the center channel's total volume; probably, you only want to do one of those things, though trying each and both is the way you are going to find out what works best in your specific case). Here are some links regarding what frequencies we are talking about:

http://en.wikipedia.org/wiki/Voice_frequency

http://en.wikipedia.org/wiki/Audio_frequency

http://en.wikipedia.org/wiki/Vocal_range

In other words, we are talking basically about the frequencies mentioned by JerryLove already in this thread.


Now, if your center channel sounds "muddy", getting a better center channel should help. Your best bet is to have your wife go with you to audition center channel speakers with dialog, preferably using the same portions of the same movies to judge this.


I personally use Aurum Cantus Leisure 2SE (original U.S. version) speakers for all channels, including the center. The sound is as clear as any theater I have ever heard (home or other). You can get a pair of the international version of these speakers for within your budget, or a used pair of the U.S. version within your budget (if you can find any for sale). However, you really should listen to speakers before you buy them, and that may be difficult with this speaker, as there are not many dealers, and probably no one you know has them.

 

[jostenmeat]

http://en.wikipedia.org/wiki/Voice_frequency

http://en.wikipedia.org/wiki/Audio_frequency

http://en.wikipedia.org/wiki/Vocal_range

In other words, we are talking basically about the frequencies mentioned by JerryLove already in this thread.

May I please ask how you derived this conclusion? Honest Q, thanks.

I don't like the idea of comparing high fidelity reproduction with the compromises of telephony. That's like saying music sounds find as low bitrate MP3 to 99.99% of the world, so therefore we should design all of our sources to be that way. Ok, that's a stretch, but maybe you get my point.

The fundamental is by far the most important component of a note being played, as IT defines the pitch.

I can't go nowhere close to 350hz without being in falsetto. I can, OTOH, sing any of the open strings on a guitar (all of which are below 350hz). Ok, I exaggerate on the lowest string, I can hit an F quietly, or F# loudly and clearly (and I have pathetic extension for a bass).

Again, I'm not saying you guys are wrong to focus on 350-3500, but why would we just toss aside the fundamental frequencies, which IMO is the most important thing?

 

[JerryLove]

I don't like the idea of comparing high fidelity reproduction with the compromises of telephony.

The good news is, as far as I can tell, that's not what we are discussing.

We are discussing the intelligible reproduction of dialog. That is exactly what a phone is designed to do. As such: it gives us a good area to look. The telephone proves that the loss of fundamental frequencies does not remove intelligibility. It may indeed harm the accuracy of voice: but the question before me is one of understanding dialog, not one of it sounding wrong.

Or maybe I missed someone's point somewhere. It's quite possible.

 

[jostenmeat]

The good news is, as far as I can tell, that's not what we are discussing.

We are discussing the intelligible reproduction of dialog. That is exactly what a phone is designed to do. As such: it gives us a good area to look. The telephone proves that the loss of fundamental frequencies does not remove intelligibility. It may indeed harm the accuracy of voice: but the question before me is one of understanding dialog, not one of it sounding wrong.

Or maybe I missed someone's point somewhere. It's quite possible.

So do you intend to mean that removing fundamentals has no ill effect whatsoever on the intelligibility of dialog? Do you think that they in fact removed fundamentals because they thought it actually improved dialogue intelligibility? Is this how I am supposed to follow your logic? Thanks.

 

[Pyrrho]

So do you intend to mean that removing fundamentals has no ill effect whatsoever on the intelligibility of dialog? Do you think that they in fact removed fundamentals because they thought it actually improved dialogue intelligibility? Is this how I am supposed to follow your logic? Thanks.

I guess you find it impossible to understand speech on a telephone.

You should click on the links I provided if you want the answer to your question. This one is short and answers your question:

http://en.wikipedia.org/wiki/Voice_frequency

Basically, any time one makes a vocalization, one produces a variety of frequencies. You do not need to hear them all in order to understand speech.

Additionally, if a frequency in your audio system is boosted, it will boost the sound of that frequency with all instruments, not just dialog. So it is a good idea to not boost any more frequencies than are needed if one is having trouble understanding dialog mixed in music and other sounds.

(Incidentally, it is because different instruments and different voices produce different additional frequencies from the main "note" that they sound different. If two things produced a sound at a particular frequency at the same volume, but with no other frequencies, they would sound identical. So the reason why a violin and a piano sound different when playing the same note is that they are also producing many other frequencies, and those other frequencies are not all the same as each other, so they sound different.)

 

[jostenmeat]

I guess you find it impossible to understand speech on a telephone.

Because I want to learn more?

You should click on the links I provided if you want the answer to your question. This one is short and answers your question:

http://en.wikipedia.org/wiki/Voice_frequency

I already did yesterday. I did again today.

Basically, any time one makes a vocalization, one produces a variety of frequencies. You do not need to hear them all in order to understand speech.

Yes, I was the first and only person to speak of harmonics before your post.

http://forums.audioholics.com/forums/showpost.php?p=773718&postcount=10

Harmonics are what defines tonal signature for us; it's what allows us to differentiate a clarinet from a violin, even when both are playing the exact same pitch.

Additionally, if a frequency in your audio system is boosted, it will boost the sound of that frequency with all instruments, not just dialog. So it is a good idea to not boost any more frequencies than are needed if one is having trouble understanding dialog mixed in music and other sounds.

Ok. I'm not sure why you are telling me this right now. Is there any particular post or passage of mine that you are addressing?

(Incidentally, it is because different instruments and different voices produce different additional frequencies from the main "note" that they sound different. If two things produced a sound at a particular frequency at the same volume, but with no other frequencies, they would sound identical. So the reason why a violin and a piano sound different when playing the same note is that they are also producing many other frequencies, and those other frequencies are not all the same as each other, so they sound different.)

Yes, right, fundamental plus the harmonics.


None of the above leads me to the conclusion that disregarding frequencies below 350hz is totally fine in achieving best dialog intelligibility.

Would I believe that 350-3500 is more important than frequencies below 350? I wouldn't be surprised, as I'm obviously seeing this figure often. BUT, does that mean sub 350 freq's are NOT important? Maybe I'm wrong, and you're sure saying I'm wrong. However, nothing you have provided has convinced me yet that I am wrong.

Again, your links have not supplied me any reasoning for the disregarding of sub 350hz. Sure, it says that the fundamental can be implied. That, to me, still does not conclusively mean that the fundamental cannot still help, or even help a lot.

Then I wonder about bandwidth constraints with telephony. Ok, so they think 350-3500 is more important, but I think of how much more this may have to do with recognizing who is on the other end of the line. Or is it because lower freq's might take up more bandwidth than upper freq's (I have no clue)? Or perhaps it's that the tiny speakers in telephones simply can't produce those lower frequencies to begin with (I have no clue). These are just questions I'm coming up with now, but again, I still haven't found anything that conclusively tells me we can disregard sub 350hz frequencies.

It just plain flat out makes no sense to me whatsoever.

While my speakers are much better with dialog intelligibility than any phone I've ever used, I would never be so stupid to completely attribute that to a better capability in reproducing vocal fundamentals. However, we ARE talking about speakers, and NOT telephones.

 

[JerryLove]

So do you intend to mean that removing fundamentals has no ill effect whatsoever on the intelligibility of dialog?

I mean that the most critical portion of the audio band for understanding speech is 350Hz-3500Hz. I

Do you think that they in fact removed fundamentals because they thought it actually improved dialogue intelligibility?

No. I believe that they kept 350Hz-3500Hz because they were most critical to understanding speech. Other frequencies are less critical.

Is this how I am supposed to follow your logic? Thanks.

You obviously may follow it however you wish. You don't seem to be following it the way I was.

Harmonics are what defines tonal signature for us; it's what allows us to differentiate a clarinet from a violin, even when both are playing the exact same pitch.

Which isn't important in comprehending the words said in dialog. To use your metaphor: the note is more important than the instrument.

Then I wonder about bandwidth constraints with telephony. Ok, so they think 350-3500 is more important, but I think of how much more this may have to do with recognizing who is on the other end of the line. Or is it because lower freq's might take up more bandwidth than upper freq's (I have no clue)? Or perhaps it's that the tiny speakers in telephones simply can't produce those lower frequencies to begin with (I have no clue).

Lower frequency speakers would have been possible. The were deemed unnecessary to the primary goal of the telephone: the conveyance of information through speech.

While my speakers are much better with dialog intelligibility than any phone I've ever used, I would never be so stupid to completely attribute that to a better capability in reproducing vocal fundamentals. However, we ARE talking about speakers, and NOT telephones.

So if you are going to try it: try it. Go into your EQ and set 350-3500HZ all the way down (and anything else you like all the way up) and listen to some conversation.

Then reverse and try again.

Remember: legibility, not pleasantness.

These are just questions I'm coming up with now, but again, I still haven't found anything that conclusively tells me we can disregard sub 350hz frequencies.

 

[jostenmeat]

I mean that the most critical portion of the audio band for understanding speech is 350Hz-3500Hz.

I'm sure there must be some AES article that has tested all this. The wiki definition of voice frequency is just lacking for me. Just how long ago was this freq range decided, back in the days of Alexander Graham Bell, and it has stuck since then? I'm sure he was smarter than me by magnitudes.

To use your metaphor: the note is more important than the instrument.

There is on argument about what is the topic: intelligibility.

There is no argument about what the topic at hand is.

Lower frequency speakers would have been possible. The were deemed unnecessary to the primary goal of the telephone: the conveyance of information through speech.

There is a distinct difference between what is or was possible, and what is done in practice. A Bose speaker has an F3 of what, 250hz? That is more than an octave above my normal speaking voice frequency. I assume that a Bose speaker/driver is larger than what is in most phones, and probably by a lot.

So if you are going to try it: try it. Go into your EQ and set 350-3500HZ all the way down (and anything else you like all the way up) and listen to some conversation.

Then reverse and try again.

Remember: legibility, not pleasantness.

NO.

Either of the below are much better comparisons.

The full range of frequencies created by the spoken voice VS Only the harmonics of said voice that reside above 350hz.

OR

The fundamental, plus ALL harmonics that are BELOW 350hz VS Only harmonics that reside above 350hz.

You are positive that both of the above will be in favor of only harmonics that reside above 350hz?

That you will always continue to tell people that dialog exists between 350-3000hz by definition as you have already done in this thread? Even when the FUNDAMENTALS for BOTH men and women are ALL below 350hz, and that's not even mentioning that the strongest harmonics for at least men will still be below 350hz? (edit: ok I see you may have now changed your definition to "critical portion".)

Bah.

I have surmised that my natural speaking voice has the fundamental (the most pronounced tone BY FAR of the various tones/harmonics that come out of body) range of 90-100hz. With harmonic series, the next most prominent will be the second harmonic, which will then lie around 180-200hz (still below 350hz). After that one, the next most prominent harmonic would be 270-300hz, again we are still below 350hz. The harmonics that would come above 350hz would all be lesser in strength than these.

I have however come across formants, which are the specific harmonics created in the vocal appartus. I am guessing it is these that might be the focus of the 350-3500 range.

Again, a question I ask is if this is more intended to recognize the individual on the other end than it does with intelligibility, as I am guessing it is the formants that allow us to distinguish who is talking on the other end of the line.

 

[TLS Guy]

May I please ask how you derived this conclusion? Honest Q, thanks.

I don't like the idea of comparing high fidelity reproduction with the compromises of telephony. That's like saying music sounds find as low bitrate MP3 to 99.99% of the world, so therefore we should design all of our sources to be that way. Ok, that's a stretch, but maybe you get my point.

The fundamental is by far the most important component of a note being played, as IT defines the pitch.

I can't go nowhere close to 350hz without being in falsetto. I can, OTOH, sing any of the open strings on a guitar (all of which are below 350hz). Ok, I exaggerate on the lowest string, I can hit an F quietly, or F# loudly and clearly (and I have pathetic extension for a bass).

Again, I'm not saying you guys are wrong to focus on 350-3500, but why would we just toss aside the fundamental frequencies, which IMO is the most important thing?

I can tell you after looking at a lot of audiograms from people with sensorineural hearing loss this is what happens.

Patients tend to present when their turnover frequency is 3.5 kHz. At a turnover frequency of 1.5 kHz they are highly symptomatic and will willingly get a hearing aid. When the turnover frequency gets to 1 kHz they are profoundly deaf as the turnover frequency drops below this point.

 

[highfigh]

Well, 3.5k hz is pretty high (think higher than the very highest fundamental that a violin can play, past the fingerboard on highest string, IIRC), but I understand that there are harmonics that make up our voices. That said, there are many men who can sing below 80hz fundamental, believe it or not.

I am not saying that 350-3500hz is not a more important range to look at (because I don't know enough about audio reproduction), but when fundamentals are concerned, it's way below 350hz.

Speech intelligibility doesn't come from the fundamentals, it's' due to the harmonics, but there is a useful limit. That's why audiologists go up to 8KHz and no higher- 9KHz- 20KHz just isn't necessary for speech. The human ear is more sensitive in the 350Hz- 3.5KHz range and this is where speech intelligibility suffers most when the person's acuity decreases.