Introduction and Spatial Localization

[KlausR.]

Hi all,

this is my first post on What’s Best so allow me to introduce myself. My name is Klaus Rampelmann, I’m examiner (petroleum engineering) at the European Patent Office, and my interest in audio was nourished by the usual hifi mags, until about 10 years ago I discovered JAES, JASA etc. Papers such as those by Lipshitz about the Great Debate and Toole about loudspeakers drastically changed my view on audio in general and high-end audio in particular.

My question relates to Spatial Mapping. From Blauert “Spatial Hearing” I understand that in natural hearing localisation blur in the horizontal plane may have values of up to 12º. I found some papers relating to localisation blur with phantom sources in 2-channel stereo:

Sandel et al., “Localization of sound from single and paired sources”, J.A.S.A. 27 (1955), no. 5, p.842

Wendt, “Directional hearing with twin-channel stereophony”, Rundfunktechnische Mitteilungen 8 (1964), no. 3, p.171 (in German)

Ortmeyer, “Localisation of sound sources in 2-channel stereophony”, Hochfrequenztechnik u. Elektroakustik 75 (1966), p.77 (in German)

The figures depend on type of test signal, frequency, listening environment (anechoic chamber, reverberation room, normal room), directivity of the loudspeakers, and there are quite substantial differences between the test subjects. Maximum value I found was 80 º off target (anechoic, 1500 Hz sine tone, no head movement).

In view of localisation blur the question is if spatial mapping is a reliable tool. What if in two test conditions the location of the sound source is perceived at two different locations within the blur zone for that test signal, that system type (2-channel, surround, etc), that type of room (anechoic, IEC, etc.): is it a real change in location or is it blur?

Related question: is there any research relating to 2-channel or surround using speech or music and what are the figures for blur?

Klaus

 

[amirm]

Welcome to the forum Klaus. I have not personally researched this area so can't provide additional references beyond what you present. Relying on others who had, level matching is super important as outlined by Sandel et al.

 

[garylkoh]

Welcome to the forum.

I'm not sure what you mean by localization blur, but this is a good paper on localization and auditory scene analysis:
http://www.cs.cmu.edu/~robust/Papers/SternWangBrownChapter.pdf

Also, a popular introduction to how we localize sound: http://www.aip.org/pt/nov99/locsound.html

 

[Kal Rubinson]

Also, a popular introduction to how we localize sound: http://www.aip.org/pt/nov99/locsound.html

That's a good one.

 

[KlausR.]

I'm not sure what you mean by localization blur, but this is a good paper on localization and auditory scene analysis:
http://www.cs.cmu.edu/~robust/Papers/SternWangBrownChapter.pdf

Also, a popular introduction to how we localize sound: http://www.aip.org/pt/nov99/locsound.html

Thanks for the links, I already had these two in my collection of papers.

Quote from Blauert: "Localisation blur is the smallest change in a specific attribute or in specific attributes of a sound event or of another event correlated to an auditory event that is sufficient to produce a change in location of the auditory event."

In other words, you don't perceive the location of the sound source where the source physically is, the apparent location is displaced with respect to the real one. According to Blauert, the absolute lower limit is about 1º, which is about two orders of magnitude less than that of the visual system, which is about one minute of an arc.

Klaus