Home Theater Forum and Systems banner

1 - 8 of 8 Posts

·
Registered
Joined
·
15 Posts
Discussion Starter · #1 ·
As many in this forum, I am in search of the best sounding audio I can achieve within financial reason – 80% solution at 20% cost. I have used room absorbers and EQ’ed the room in which I will be the only listener – and am almost there. As this is probably a perpetual process, the advancements in programmable digital electronics are extremely exciting and leave me wanting more towards this goal – however I do not buy into hype and use my ear as the final determinant. So with the large knowledge base in this forum, I have a couple questions that I hope someone can help me with:

1) Onkyo incorporates a dynamic loudness that, in my opinion, is extremely effective – it may be part of the Audysses system. This adjusts the bass and treble according to the volume music is played at. Is there an add-on solution that anyone knows of that can be applied to an older 2.1 channel system? … an add-on that follows the “Loudness vs Frequency” curves on page 46 (Fig. 4-6) in the 5th ed. of Master Handbook of Acoustics, for example? I would think it relatively straight forward to implement.

2) Samsung/Android has an app for its “Music Player” called “Adapt Sound.” The app goes through a routine that adjusts the frequency response of the headphone output to your hearing by asking if you hear various tones in either ear. I find that this is also extremely effective and I am wondering if this kind of thing is offered in a standalone DSP – MiniDSP for example?

Any help would be greatly appreciated!
 

·
Banned
Joined
·
4,838 Posts
Is there an add-on solution that anyone knows of that can be applied to an older 2.1 channel system? … an add-on that follows the “Loudness vs Frequency” curves on page 46 (Fig. 4-6) in the 5th ed. of Master Handbook of Acoustics, for example? I would think it relatively straight forward to implement.
I am not sure which set of curves is referred to above, the older Fletcher-Munson curves, or a more recent iteration like the ISO 226:2003 Equal Loudness Contours which are supposedly more accurate and have higher bass levels. That might not be super critical. The implementation could be quite involved, with
  • a frequency response curve that is constantly changing with level
  • gain constantly changing with level
  • attack, sustain, and decay time constants
Sounds doable but fairly involved. I have not heard of a DSP with a developer's toolkit for those functions, although there could be something out there.

Samsung/Android has an app for its “Music Player” called “Adapt Sound.” The app goes through a routine that adjusts the frequency response of the headphone output to your hearing by asking if you hear various tones in either ear. I find that this is also extremely effective and I am wondering if this kind of thing is offered in a standalone DSP – MiniDSP for example?
I have seen the apps. They work at the hearing threshold, and regular listening level involves a different curve, does it not? Wouldn't that cause some inaccuracy? Again, maybe not super critical for what you are trying to accomplish. Have seen nothing for a DSP, though.
 

·
Registered
Joined
·
15 Posts
Discussion Starter · #3 ·
AudiocRaver, thanks for your response. The set of curves I am referring to come from (according to the bibliography in the Master Handbook: Robinson, D. W. and R. S. Dadson, “A Re-Determination of the Equal-Loudness Relations for Pure Tones,” British J. Appl. Psychology, 7, pp. 166-181, 1956.). They are similar to the Fletcher-Munson curves you point out, but not the same (http://www.sengpielaudio.com/Fletcher-MunsonIsNotRobinson-Dadson.pdf). Any of them, including the ISO you mention, would surely make low volume listening sound better.

I think what you are discussing is more complex than I imagined. In what I am looking for, the curve would change according to the state of the volume knob, for example (or some long time averaged output, or equivalent) – it is not a compression scheme. I am not sure how the system in the Onkyo’s are implemented, but I would guess it is somewhere between a simple loudness button and what I discuss.

Thus, I would think it straight forward to implement – simply different equalization according to the quasi-static output level or volume knob. – probably nothing more than a firmware routine for the MiniDSP.

The Adapt Sound would be similar, and it would truly personalize the audio (for 1 listener anyway). As a person ages, it can be readjusted according to their hearing change. This is something that should be incorporated in hearing aids btw so that a user can adjust it (not going any deeper into that can of worms :)).
 

·
Banned
Joined
·
4,838 Posts
AudiocRaver, thanks for your response. The set of curves I am referring to come from (according to the bibliography in the Master Handbook: Robinson, D. W. and R. S. Dadson, “A Re-Determination of the Equal-Loudness Relations for Pure Tones,” British J. Appl. Psychology, 7, pp. 166-181, 1956.). They are similar to the Fletcher-Munson curves you point out, but not the same (http://www.sengpielaudio.com/Fletcher-MunsonIsNotRobinson-Dadson.pdf). Any of them, including the ISO you mention, would surely make low volume listening sound better.

I think what you are discussing is more complex than I imagined. In what I am looking for, the curve would change according to the state of the volume knob, for example (or some long time averaged output, or equivalent) – it is not a compression scheme. I am not sure how the system in the Onkyo’s are implemented, but I would guess it is somewhere between a simple loudness button and what I discuss.
Leave it to me to think of the most complex possible approach. What you are suggesting would be far more practical, and might be approximated with a number of parametric filters, commonly available with DSP units. Maybe four or five approximated curves, selected to cover a volume range. How to activate/select is the remaining question.

You might look at REW's target curve function as a way to generate filter values, see what it comes up with.

Thus, I would think it straight forward to implement – simply different equalization according to the quasi-static output level or volume knob. – probably nothing more than a firmware routine for the MiniDSP.

The Adapt Sound would be similar, and it would truly personalize the audio (for 1 listener anyway). As a person ages, it can be readjusted according to their hearing change. This is something that should be incorporated in hearing aids btw so that a user can adjust it (not going any deeper into that can of worms :)).
Interesting idea.
 

·
Registered
Joined
·
15 Posts
Discussion Starter · #5 ·
Great! I have not found the target curve function in REW – only “target settings” in the EQ section which might be too limiting – no way to add gain in the bass/treble or cut midrange. Am I missing it?

I don’t have a MiniDSP but might get one soon – do you know if the software is open architecture? I might write something myself but I don’t want to reinvent the hardware wheel. Apparently it has a digital volume access point which could be utilized as the control.

As far as I am concerned, over thinking a problem is an optimization process which is always a good thing :)
 

·
Banned
Joined
·
4,838 Posts
Oops, meant to say House Curve. It is on the House Curve tab under Preferences. There are instructions for file format on that page. You will need a fake measurement to apply correction to. Import a saved measurement file into an Excel spreadsheet, and change all of the SPL values to the same number, say 80 dB, or whatever, all phase to 0 deg. Then you can generate filters for that curve and the House Curve will be applied.
 

·
Registered
Joined
·
15 Posts
Discussion Starter · #7 ·
Excellent idea AudiocRaver! Thanks for the details - got it – and also the original measurement file has to be exported as a text file from REW instead of saved. I will play around with this a bit and see where it leads.
 
1 - 8 of 8 Posts
Top