An Exploratory Study of Western Orchestration: Patterns through History

Abstract

Changes in instrument combination patterns in Western classical orchestral music are traced over a three hundred year period from 1701 to 2000. Using a stratified sample of sonorities from 180 orchestral works by 147 composers, various empirical analyses are reported. These include analyses of instrumentation presence, instrument usage, ensemble size, common instrument combinations, instrument clusterings, and their changes over time. In addition, the study reports associations of different instruments with various dynamic levels, different tempos, pitch class doublings, affinities between instruments and chord factors, as well as interactions between pitch, dynamics, and tempo. Results affirm many common intuitions and historical observations regarding orchestration, but also reveal a number of previously unrecognized patterns of instrument use.

Can Orchestration Teaching be improved?

“…Chon, Huron, and DeVlieger [5] summarized the most frequent combinations found in orchestral repertoire from 1700 to 2000. This information can be very useful in music analysis to generate expectations and distinguish between the normative and the extraordinary use of instruments.…”

CNN based musical instrument identification using time‐frequency localized features

“…Identification of instruments from music excerpts has varied applications in sound engineering 1 and musicology. 2 Proper identification of instruments is required for musical source separation 3 which facilitates different forms of sound engineering such as mixing, karaoke or, instrument-wise equalization. 1 In musicology, identification of individual instruments in orchestral music is important to understand how presence, usage and combinations of instruments evolved with time.…”

“…1 In musicology, identification of individual instruments in orchestral music is important to understand how presence, usage and combinations of instruments evolved with time. 2 The identification task, however, belongs to one of the most challenging areas in music information retrieval. The fine differences between two different instruments can only be recognized by veterans of the domain 4 and often goes unattended to the lay man and even to many algorithms that have been applied to the problem.…”