Speech & listening quality

Plans and things, layout, style, where do I put my near-fields etc.

Moderators: Aaronw, kendale, John Sayers

DanMilo
Posts: 13
Joined: Fri Dec 21, 2012 3:43 am
Location: Germany

Speech & listening quality

Post by DanMilo »

Hi,
hope you could give me some tips here.
My goal is to improve speech and listening quality in a room of 3,5m x 4,3m x 2,5m. The walls are stone and sheetrock with wallpaper on it. The floor is a short carpet. I layed 4m2 laminat on floor to have a more linear t60 in the upper end because the room seems really dead with furniture in it. The actual t60 is 0,28-0,3.
The room is used for talking and listening to music on 2 or 3 different small speaker systems. It should not be muddy or boxy. At the moment it is really boxy in the 100-400hz range I think. Not nice to talk or listen to music.
I tried to install 2 x 1m2 of basotect (25cm thick) in different location, but the room was totally dead and awkward. I think what makes it awkward could be peaks from axial modes or the rining from those modes and the loss of more high end frequency. What do you think? Where is the best position to place the absorbers to have the best effect in general against the mud. Would it be centered on the walls to dampen the axial modes...or should it be the corners? And if I measure different locations in the room....after what should I look (the ringing) or more the avareged frequency response? I´m planning to install some wood or diffusors on the absorber if needed to give some reflections back. Thank you :wink:
DanMilo
Posts: 13
Joined: Fri Dec 21, 2012 3:43 am
Location: Germany

Re: Speech & listening quality

Post by DanMilo »

Perhaps the question is a little bit too confusing?

In a direct comparison, what is more effective to bring down strong axial modes or to give better speach quality when i only have 4 porous traps? Place them in the corners or at the center of the walls so they ould also catch some sbir? I know, that corner mounting works more general. But i´m not sure if the axial modes would be more effectivly damped at the center. thank you :shot:
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Speech & listening quality

Post by Soundman2020 »

The actual t60 is 0,28-0,3.
For a room used for speech, that's a little low, but not by a huge amount. It should probably be more in the range 0.4 - 0.45 for a room that size. It would seem that there's just too much absorption overall, and not enough reflection/diffusion. But it might also be the case that the absorption just isn't balanced well across the frequency spectrum.

Without seeing actual acoustic test data from the room, it's hard to give you advice on what to do since we don't know what the problems actually are! If you could run a REW test in there and post the results, that would help.
The room is used for talking and listening to music on 2 or 3 different small speaker systems.
Is this for some type of speaker evaluation system? Or just a music listening room? Also, optimizing for music and also for speech isn't that easy, as there are different acoustic recommendations for each. If you could be more specific about what the room will be used for, and also provide some photos, that would help too.
I think what makes it awkward could be peaks from axial modes or the rining from those modes and the loss of more high end frequency.
That's possible, but getting actual acoustic data would be the only way to be certain. A "boxy" sound is usually due to over-absorption of one frequency range, and/or a lack of reflection/diffusion in another frequency range.

Set up a REW test with the mic in the normal listening position, and using the best speakers that you would normally use in there, set up in their normal positions. The draw a diagram of where all those are in relation to the walls, floor, and ceiling, and post that too.
Where is the best position to place the absorbers to have the best effect in general against the mud.
Corners. Corners are always best, because you get a free "boost" in effectiveness, and you get to deal with the modes from two axes at once, or all three axes if you are covering the tri-corners well.
Would it be centered on the walls to dampen the axial modes...
Axial modes aren't just related to the center of the wall: they are related to the entire wall, from corner to corner, and floor to ceiling. And since all modes terminate in the corners, by putting the trapping in the corners you are assured of hitting all the modes, not just the axial ones.

However, using lots of thick absorption for your bass traps will also kill the high end of the spectrum, and the mids too. You need to compensate for that, by using selectively tuned foils in front of the absorption. The foil will reflect the high frequencies back into the room, to keep it lively, while still allowing the lows through to be damped by the absorption.
And if I measure different locations in the room....after what should I look (the ringing) or more the avareged frequency response?
Both! :) In fact, look at everything! There's a lot of acoustic things happening all at once in any room, and the overall psycho-acoustic impression that your brain receives is a combination of all of them. First and foremost is to ensure that your modal issues are damped as much as reasonably possibly, but you also need to be looking at SBIR issues (which can have a big effect), as well as flutter echo, diffusion, specular reflections, frequency response, etc. Possibly the most important view is the ETC data, provided that you understand how to interpret it. But look at everything, then treat the biggest issues one by one, but also making sure that you are not causing other issues in the process.
I´m planning to install some wood or diffusors on the absorber if needed to give some reflections back.
Don't do that until you understand what the problems are! If you don't know what you are dealing with, then how would you decide what frequency to tune the diffusers too? :)

First do the test, then identify the problems, then decide how to deal with each of them, one by one, or in related groups.
what is more effective to bring down strong axial modes or to give better speach quality
Speech isn't usually affected by strong axial modes. The frequencies are totally different. Axial modes are usually a problem in small rooms up to about 200 Hz, but mostly below around 100 Hz. There's very little useful energy in speech at such low frequencies. Yes, there are some fundamentals down there, but actual intelligibility is much higher up the spectrum. The problem with speech intelligibility is usually more in the mid range and the lower part of the highs.


- Stuart -
DanMilo
Posts: 13
Joined: Fri Dec 21, 2012 3:43 am
Location: Germany

Re: Speech & listening quality

Post by DanMilo »

Okay, thank you so much, for taking the time.
I´ll post some measurements.
I have no absorber in the room yet. There are 3 small desks....with a pair of speaker on each of them and a small system in the shelf at the back of the room. I have now covered 4qm2 with vinyl floor over the carpet. The rt60 looks better in the upper end when measuring at different spots in the room, but the walls have a soft wallpaper attached. Perhaps these absorb a little to much of the high end, too? Speach sounds a tiny tiny bit boomy and bassheavy in the room, but only a tiny bit . Music also. When i bring in a bastoct absorber 50cmx100cm or a glaswool or mineralwool of 15cm thickness the sound in the room changed a lot and gets muddy and its not nice to speak. When wrapped in thick foil it gets a little bit better, but it still is worses than without an absorber in the room. I tried different corners (bottom, tricorner, but i have the feeling that the room gets a kind of pillow awkward sound when sitting there for some hours). I attached even some thick pvc foil or thin foil....when i bring in more absorber in different corners it gets really really unnatural. So i think it could be from three problems (modal, sbir, rt60). But the evidence it gets worser with absorption directs more to rt60? I´m not sure...it sounds like i´m sitting in a big pillow when i bring it absorption.
DanMilo
Posts: 13
Joined: Fri Dec 21, 2012 3:43 am
Location: Germany

Re: Speech & listening quality

Post by DanMilo »

Okay, here are some measurement from one desk with different speakers. The bigger peak at 120 seems to be from a resonance between desk and the ceiling,...when i place the absober above the desk it looses energy. But the overallroomsound gets even more pillow...measurements are without absorber.
Sad that i cant upload the mdat.
http://expirebox.com/download/e2cdbad76 ... 29918.html
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Speech & listening quality

Post by Soundman2020 »

When wrapped in thick foil it gets a little bit better,
What foil? How thick? You can't just use anything! You have to calculate the thickness, according to the type of foil and the frequency range that needs treating.
in the shelf at the back of the room.
Shelf??? what shelf??? You didn't mention any shelf. You said the room was empty...
I have no absorber in the room yet.
There must be something in there that is acting as general broadband absorber. If not, the room would be much more live than it is! For that size room with no treatment, just solid walls and ceiling (drywall) plus carpet on the floor, the RT60 time would be around 600 ms. With no carpet, it would be over 1200 ms. In your case, it as around 250, and fairly even across the spectrum. Therefore there is something that is acting as treatment.
The rt60 looks better in the upper end
It's no use looking at just the RT60, and just a part of the spectrum. That doesn't tell you much about the room. RT60 is just one small aspect of room acoustics, and isn't even the most important. The overall RT60 graph for that room is not too bad, in all three cases.
when measuring at different spots in the room
The only places where you need to measure are those specific locations where critical listening occurs.
the walls have a soft wallpaper attached. Perhaps these absorb a little to much of the high end, too?
Unlikely. It would have to be VERY thick to have a substantial effect, except at the very high end.
When i bring in a bastoct absorber 50cmx100cm or a glaswool or mineralwool of 15cm thickness the sound in the room changed a lot and gets muddy and its not nice to speak.
If adding absorption makes the room sound muddy, then the absorption is no good for what you are trying to do. It is absorbing too much of the mid range, and not enough of the low frequencies.
But the evidence it gets worser with absorption directs more to rt60?
You seem to be far too worried about RT60. In acoustics, overall RT60 is not the most important aspect: it is just one of many acoustic aspects that all combine to produce the final sensation of the room. If your room sounds worse ("muddy") when you add absorption, that would indicate that the absorption you are adding is not of the correct type, or the device is not built correctly, or not positioned correctly.

I looked at your REW data,but either you didn't calibrate REW correctly, or you did the tests at a level that is about one thousand times too low. REW shows that you did the tests at about 55 dB, but that should be about 85 dB. Did you calibrate REW correctly, using a proper hand-held sound level meter? If not, then you need to do that, then repeat the tests. You also did not run the test at low enough frequencies: you missed out an entire octave! You started at 40 Hz, but we need to see down to the bottom end of the spectrum: start your test at 18 Hz, and end it at 22,000 Hz.
The bigger peak at 120 seems to be from a resonance between desk and the ceiling,...when i place the absober above the desk it looses energy.
That is almost certainly a modal issue, not flutter echo. It is strongly present in all three tests, and is close to the frequency for your 3,0,0 mode (3rd harmonic, axial, length). The other very clear modal issue is the one at about 49 Hz, which is obviously your fundamental 0,1,0 mode (axial, width).

There's a general lack of mid-range, and also extreme highs, but that could be due to the equipment setup rather than the room.

I can't say much more than that, since you didn't give me any of the other information that I asked for.


- Stuart -
DanMilo
Posts: 13
Joined: Fri Dec 21, 2012 3:43 am
Location: Germany

Re: Speech & listening quality

Post by DanMilo »

:D thank you....try to answer as best as i can.
-Foil: i tried wrapping film (seems not to do much), rubbish bag of 120L (thicker), synthetic leather (pvc)
-Furniture: 1bigger desk, 2 small desks, a hanging shelf at the backwall with some books and a ghettoblaster (like this one Image), a 1m cat tree, 2barstool
-Absorber: On the backside of the door hang two jackets, the carpet on the floor, wallpaper except the ceiling, 2 chairs with fabric, the window has a small alumunium rollo attached that diffuses a little bit, open door into the floor that is really bright with laminat, thats it. When i take of the jackets and raise the rollo the room gets more life. Thats much better for speech quality, never thought that this small change made that much of a difference!!! Its now at 0,32sec.

Okay,...so you think i should make the absorber bigger to get lower frequencies or change the material? I tried basotect and not so dense glaswool. Basotect is not so good in the lowfrequ absorption....but its a 10cm thick panel (have 3 of them to work with). When i bring the basotect into the room its getting really pillow again. I thought of using these panels as corner absorber. But when i place them in any corber the overall acoustic gets damped. So i think i´ll try some bigger mineralwool pallets covered with material that they dont suck out the high end in the room again.

BTW. i used a cal file for the mic. The spl wasnt calibrated but i measured nearly 80db.
Yes, could be the 3 length mode....

Thank you Stuart!
Soundman2020
Site Admin
Posts: 11938
Joined: Thu Aug 21, 2008 10:17 am
Location: Santiago, Chile
Contact:

Re: Speech & listening quality

Post by Soundman2020 »

Please provide some photos of the room. I can't help you if I don't understand the room.

- Stuart -
Post Reply