Certainties and Uncertainties from Using a Selection of Data to Predict Concert Hall Preference. Issue 4 (December 2013)

• Record Type: Journal Article
• Authors: Skålevik, Magne
• Is Part Of: Building acoustics. Volume 20:Issue 4(2013)
• Page Start: 335
• Page End: 350
• Publication Date: 2013-12
• Subjects: Soundproofing -- Periodicals; Architectural acoustics -- Periodicals; Acoustical engineering -- Periodicals; Acoustical engineering; Architectural acoustics; Soundproofing; Periodicals; 693.834
• DOI: 10.1260/1351-010X.20.4.335
• Languages: English
• ISSNs: 1351-010X

Abstract

Over the past hundred years or so, many researchers have explored the possible correlation between physical properties of the concert halls and listeners assessment of the acoustics of the same halls. And we are still searching and researching. This author has previously shown how some sets of room acoustical parameters can, with their appropriate qualifying criteria, be used to explain the subjective ranking of a selection of halls from Beranek's rank ordering of 58 halls. A set of five listening aspects in ISO-3382 seems to be important, but trials with even more physical quantities have provided more explanation potential. A critical limitation in the research turned out to be the lack of sufficient amount of subjective AND objective data, leading to the launch of an online concert hall acoustics rating survey. In this paper, the latest results from this author's investigation are presented, featuring a demonstration of how the size of selected data affects the statistical uncertainties in such results. Remaining uncertainty in the prediction method naturally leads to a "safety first" policy with strict acceptance limits for the objective data. As a consequence, many appreciated halls would not be recommended for replication. These and other consequences need to be discussed in further work.