Pitch content on historic field recordings

Analyzing 400 wax cylinder recordings

Joren Six - IPEM, Ghent University - Belgium
Olmo Cornelis - Indiana University Bloomington - USA
joren.six@ugent.be
Feb 2023

Patterns (in music)




“Patterns are fundamental in music”
Bod, R. (2022). World of Patterns: A Global History of Knowledge. JHU Press.

Relevance

Why study patterns in music?

  • Origins of music
  • Evolution of music
  • Non-human musicality
  • Nature-nurture debates
  • Definition of music

Brown, Steven, and Joseph Jordania. "Universals in the world’s musics." Psychology of Music 41.2 (2013): 229-248.

Study of patterns in music

Action and Perception
  • Context-poor
  • Data-poor
  • Controlled
  • Many studies
Corpora
  • Context-rich
  • Data-rich
  • Wild
  • Few large scale

Study of patterns in music

Corpora
  • Context-rich
  • Data-rich
  • Wild
  • Few large scale
Savage, P. E. (2022). An overview of cross-cultural music corpus studies.

Panteli, M., Benetos, E., & Dixon, S. (2018). A review of manual and computational approaches for the study of world music corpora. JNMR.

Study of patterns in music

Why now? Two major shifts in the last decades:

  • Availability of large music archives in digital form
  • Advances in Music Information Retrieval (MIR)
Panteli, M., Benetos, E., & Dixon, S. (2018). A review of manual and computational approaches for the study of world music corpora. Journal of New Music Research, 47(2), 176-189.

Study of patterns in music

Potential patterns common in music around the world:

  • Distinctness of pitches
  • Octave equivalence
  • Number of pitch classes
  • Intervals between pitch classes
Brown, Steven, and Joseph Jordania. "Universals in the world’s musics." Psychology of Music 41.2 (2013): 229-248.

Methods: pitch tracking

Methods: pitch tracking

Collection

  • Culturally d iverse
  • Geographically spread
  • 'Uninfluenced'


  • Noisy
  • Short
  • Unbalanced

Fig: Wax Cylinders, most poplar around 1896–1916 with a capacity of about 2 minutes.

Collection

IU - K.E. Laman, 1911, French Equatorial IU - G. Herzog, 1930, Liberia RMCA - P. Tempels, 1944, RDC RMCA - P. Tempels, 1944, RDC - Denoised
Moliner, E., & Välimäki, V. (2022, May). A two-stage u-net for high-fidelity denoising of historical recordings. ICASSP 2022-2022. IEEE.

Collection


Fig: Conceptual problems with applying deep learning models on an out-of-domain corpus.
Moliner, E., & Välimäki, V. (2022, May). A two-stage u-net for high-fidelity denoising of historical recordings. In ICASSP 2022-2022. IEEE.

Collections

Collection Nr Institute Date Collector Country Number Analyzed
59.002 IU 1900-1913 (Berlin archive) Various 43 32
69.015 IU 1905 F.A. Starr Zaire 68 8
MR.1959.8.8 RMCA 1910 G. Hulstaert RDC 31 45
MR.1959.8.9 RMCA 1910 G. Hulstaert RDC 31 45
60.019 IU 1910 Speck Nigeria 20 5
MR.1959.5 RMCA 1912 A.Hutereau RDC 254 201
54.21 IU 1913-1922 K.E. Laman Congo 7 7
54.211 IU 1913-1922 K.E. Laman French Equatorial 21 11
54.223 IU 1930-1931 G. Herzog Liberia 236 0
67.152 IU 1931 F. Herkovits Ghana/Nigeria 242 5
MR.1959.6 RMCA 1934 Unkown RDC 72 54
54.003 IU 1935 R. H. Morey Liberia/SL 14 5
MR.1961.5.15 RMCA 1944 P. Tempels RDC 41 5

Preliminary findings

  • The concept of pitch class is present
  • 170 or 240 cents as building blocks
  • 2 to 8 pitch classes
  • Fifth is almost always present.
Fig: The number of songs for each pitch class set size.

Preliminary findings

Fig: Pitch intervals between all pitch classes for recordings with 5 identified pitch classes.

Preliminary findings

Fig: Pitch intervals between all pitch classes for recordings with 4 identified pitch classes.

Discussion

  • Bias: selection or recording technology
  • Analysis assumes octave equivalence (498 = 702)
  • Absolute pitch unclear
  • Timbre ignored
  • Unbalanced dataset
  • Release dataset without audio
  • Separating scale origins difficult
    • perception
    • production
    • information content minimization
    • transmission
McPherson, M. J., & McDermott, J. H. (2023). Relative pitch representations and invariance to timbre. Cognition, 232, 105327.

Further reading

0110.be
  • Six, J., Cornelis, O., & Leman, M. (2013). Tarsos, a modular platform for precise pitch analysis of Western and non-Western music. Journal of New Music Research, 42(2), 113-129. Retrieval.
  • Chenot, J. H., Laborelli, L., & Noiré, J. É. (2018). Saphir: optical playback of damaged and delaminated analogue audio disc records. Journal on Computing and Cultural Heritage, 11(3), 14-1.
  • Savage, P. E. (2022). An overview of cross-cultural music corpus studies.
  • McPherson, M. J., & McDermott, J. H. (2023). Relative pitch representations and invariance to timbre. Cognition, 232, 105327.
  • Moliner, E., & Välimäki, V. (2022, May). A two-stage u-net for high-fidelity denoising of historical recordings. In ICASSP 2022-2022. IEEE.
  • Brown, Steven, and Joseph Jordania. "Universals in the world’s musics." Psychology of Music 41.2 (2013): 229-248.
  • Bod, R. (2022). World of Patterns: A Global History of Knowledge. JHU Press.
  • Phillips, E., & Brown, S. (2022). Vocal imprecision as a universal constraint on the structure of musical scales. Scientific Reports, 12(1), 19820.

Preliminary findings

Fig: Pitch intervals between all pitch classes for recordings with 3 identified pitch classes.

Preliminary findings

Fig: Pitch intervals between all pitch classes for recordings with 6 identified pitch classes.

Preliminary findings

Fig: Pitch intervals between all pitch classes for recordings with 7 identified pitch classes.