There are indications that humans are only able to correctly count up to three voices in polyphonic music pieces of homogeneous timbre, where each voice is played by the same instrument. A more general case, where voices are played by instruments of inhomogeneous timbre, has not been fully addressed so far. In order to approach this question we conducted a listening experiment with 62 participants to find out whether both scenarios - instrumentation by inhomogeneous or homogeneous timbre - share the same outcome. This paper describes the design of the experiment including an analysis of the results, which show that both scenarios are related. Furthermore, a detailed analysis of the error rates in correctly counting the number of instruments reveals that there are significant differences between non-musician and musician listeners, in particular regarding the upper auditory limit of the number of correctly counted instruments. Based on these results, models for the perception of instruments in auditory str...
[1]
Stephen McAdams,et al.
Hearing Musical Streams
,
2008
.
[2]
Tatsuya Hirahara,et al.
How many concurrent talkers can we hear out
,
1996
.
[3]
David Huron.
Voice Denumerability in Polyphonic Music of Homogeneous Timbres
,
1989
.
[4]
T. Jaeger,et al.
Categorical Data Analysis: Away from ANOVAs (transformation or not) and towards Logit Mixed Models.
,
2008,
Journal of memory and language.
[5]
Guy J. Brown,et al.
Computational Auditory Scene Analysis: Principles, Algorithms, and Applications
,
2006
.
[6]
Masataka Goto,et al.
RWC Music Database: Popular, Classical and Jazz Music Databases
,
2002,
ISMIR.
[7]
Morten W Fagerland,et al.
The Wilcoxon–Mann–Whitney test under scrutiny
,
2009,
Statistics in medicine.
[8]
E. Cambouropoulos.
Voice And Stream: Perceptual And Computational Modeling Of Voice Separation
,
2008
.