[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Critical bands v hearing loss

MPEG sound compressors work by coding spectra on a critical
band basis, and they've been quite successful. However,
the compressed sounds often aren't perceptibily identical
to the originals. So it's a matter of how much inaccuracy
is acceptable for the particular application.


On Wed, 28 Sep 2005, Richard H. wrote:

> In the radio world, using radio reflectors etc sized at 10% of the wavelength is regarded as precise enough for accurate operation.
> I wonder if, in hearing aids, using more "bands" each set to say 10% or so of the critical band width would ensure "precise" processing?
>   ----- Original Message ----- 
>   From: Jan Schnupp 
>   To: AUDITORY@xxxxxxxxxxxxxxx 
>   Sent: Wednesday, September 28, 2005 9:20 AM
>   Subject: Re: Critical bands v hearing loss
>   Jim is right to point out that things aren't as simple as that. You could play a 100 Hz tone and a 110 Hz tone, but slightly offset their interaural levels and their onsets, and they will be perceived as two separate tones even though they fall into the same critical band. Or you can play a 100Hz tone and a 300Hz tone simultaneously with the same interaural level and you will perceive only one tone (albeit a "complex", "harmonic" one) even though you now excite two well separated critical bands. I wouldn't claim to be an expert, but the impression I have is that, when it comes to auditory scene segmentation, critical bands are not that critical. Or am I wrong there?
>   Jan
>   beaucham wrote:
> On Thu Sep 22 01:26:43 2005 Vinay S.N wrote:
>   The human ear can only make sense of one signal per critical band. For
> example, if two sine tones of 100Hz and 110Hz are played, two distinct
> tones are not heard. This is because the critical bandwidth at 100Hz is
> about 80Hz, and the two tones would be within the same critical band; one
> will only hear two distinct tones when the two excitations exist in
> separate critical bands. Hence there is no point in having controllable
> bands which are finer in resolution  than a critical band.
> I don't think this is true in general. If the tones are too close
> together (less than 20 Hz) beats will be heard. Between 20 and 40 Hz
> difference a roughness sensation is heard. Beyond that two tones can
> be heard. Also, if there are three tones, the relative phases can be
> aligned to imitate amplitude OR frequency modulation, and they sound
> entirely different. The question is how much detail can be heard
> within a critical band. The case I referred to in my previous note
> was the harmonic spectrum case, which I neglected to mention.) I'm
> sure this question has been explored, and I'd like to see some
> discussion on this topic by people in the psychoacoustic area.
> Jim