Organizations: Sony Computer Science Laboratories, 6 rue Amyot, 75005 Paris, France · Department of Architecture, Design and Media Technology, Aalborg University, Rendsburggade 14, 9000 Aalborg, Denmark · Neodrome Entertainment, 5 rue Vernet, 75008 Paris, France · Independent performing musician, Senjak, Belgrade, Serbia · Citizen Records, 2 rue Jacques Cellerier, 21000 Dijon, France
This study argues that electronic tones routinely used in contemporary popular music - including 808-style bass and power chords - are structurally and perceptually equivalent to multiphonics in contemporary classical music. Using listening tests (n=10) and signal analysis, we show that both types of tones elicit multiple, listener-dependent pitch percepts arising from similar spectral and temporal features. These findings suggest that pitch ambiguity is not confined to experimental classical contexts but is also a feature of mainstream music production.
Figures & tables
Sample
Section
Production
Source
Number
method
of perceived
pitches
Cello F2 tone
4
Acoustic
This study
1.64
Cello natural harmonic
5.1
Acoustic
This study
1.66
Horn multiphonic
5.2
Acoustic
Weber ( 2012 )
1.86
Oboe multiphonic
6.3
Acoustic
Holliger ( 1971 )
1.95
Table 1: Samples used in the study.
Figure 1: Cello tone, F2 on the C string, weighted audio. Power spectrum. The dot’s size reflects the energy. The x-axis shows multiples of the f0 .
Figure 2: Cello tone, F2 on the C string, listening test results. In white, pitches for which d=0 ; in green, d=1 ; in orange, d=2 ; in red, no association with the signal. The same protocol will be used in the subsequent figures. The bar graph to the left of the pitch-tracker results compiles the perceived pitches together with their associations. The label “h5-2”, for example, indicates an association with harmonic 5, two octaves below. Here, C5 is not associated with harm. 6 but is described as one octave below C6 (harm. 12), as the C6 partial is salient in Fig. 1 .
Figure 3: Cello recordings, weighted audio. Power spectrum peaks corresponding to F2 (fundamental), F2+C4 (multiphonic), and C4 (harmonic 3). The dots highlight the partials corresponding to C4. The x-axis shows multiples of the F2 f0 .
Figure 4: (a) F2+C4 and (b) C4 tones from Fig. 3 , listening test results. In (b), “h27”, the 27 th harmonic of F2, is the 9 th harmonic of C4.
Figure 5: Fallowfield’s “AudioEx11 III 3–5–8–13 EK” sample ( Fallowfield, 2019 ) , weighted audio. Power spectrum. Harmonic tones with fundamentals corresponding to partials 1, 3, 5, 8, and 13 are shown in blue, yellow, red, green, and white, respectively.
Figure 6: Fallowfield’s tone from Fig. 5 , listening test results.
Figure 7: Weber, Concertino in E, bar 172 multiphonic, weighted audio. Power spectrum. The vertical line shows the predicted pitch value according to temporal modeling (B1).
Figure 8: Weber, Concertino , tone from Fig. 7 , listening test results. The association between D#1 and harmonic 5 is doubtful.
Figure 9: Metallica, … And Justice for All , 1’07–1’14, extracted guitar track (one power chord), weighted audio. Power spectrum. The vertical line shows the predicted pitch value according to temporal modeling (E1). The x-axis shows multiples of the f0 .
Figure 10: Metallica, … And Justice for All , tone from Fig. 9 , listening test results.
Figure 11: Vitalic, See the Sea (red) , 0’50–0’52, extracted bass track, weighted audio. Power spectrum. The x-axis shows multiples of the f0 .
Figure 12: Vitalic, See the Sea (red) , tone from Fig. 11 , listening test results.
Figure 13: Heinz Holliger, Studie über Mehrklänge für Oboe Solo , first “note”, weighted audio. Black, power spectrum. Yellow, power spectrum, smoothed. Blue box, predicted pitch value according to temporal modeling, of which the x-axis ticks are multiples. White box, same value but for the smoothed spectrum.
Figure 14: Holliger, Studie über Mehrklänge , tone from Fig. 13 , listening test results. The harmonics correspond to the f0 of the carrier (smoothed power spectrum). The pitches in black correspond to individual partials that are close to, but distinct from, the f0 .
Figure 15: Primaal, ¡Fire! , bass, weighted audio, one frame. Black, power spectrum. Yellow, power spectrum, smoothed. Blue box, pitch value predicted from temporal modeling.
Figure 16: Primaal, ¡Fire! , tone from Fig. 15 , listening test results. The association between D6 and harm. 37 (D7) stems from the high energy of this partial. Low-register pitches around C/C#/D are merged in the bar graph.
Figure 17: Primaal, Silver , bass, resynthesized bass frame, weighted audio. (a) Power spectrum. The x-axis ticks are multiples of the first peak’s frequency (in blue). (b) Frequency differences between consecutive partials. The blue line denotes twice the frequency of the first peak.
Figure 18: Primaal, Silver , bass, tone from Fig. 17 , listening test results. Pitches around F# are merged in the bar graph. D2 is expressed as “ f0 +1” as the tone features only odd harmonics.
Figure 19: Walter, multiphonic from Versunkene Form , weighted audio. (a) Power spectrum. White box, predicted pitch value according to temporal modeling. (b) Frequency differences between consecutive partials.
Figure 20: Walter, Versunkene Form , tone from Fig. 19 , listening test results.
Figure 21: Vitalic, No Fun , main synthesizer part, last “note” of the main pattern, resynthesized sample, weighted audio. Power spectrum. White box, the predicted pitch value according to temporal modeling, of which the x-axis ticks are multiples. The partials underlined in pink form a quasi-harmonic tone.
Figure 22: Vitalic, No Fun , tone from Fig. 21 , listening test results. The approximate pitch corresponding to each partial’s frequency is indicated in place of the harmonic number.
Aims. This study suggests that the use of multiple perceived pitches arising from a single harmonic complex tone is an active and intentional feature of contemporary popular music. The phenomenon is illustrated through examples drawn from the work of electronic artist Vitalic and others. Methods. Two listening tests were conducted: (1) evaluation of the number of simultaneous pitches perceived from single harmonic tones, and (2) manual pitch transcription of sequences of harmonic tones. Relationships between signal characteristics and pitch perception were then analyzed. Results. The synthetic harmonic tones found in the musical sequences under study were observed to transmit more perceived pitches than their acoustic counterparts, with significant variation across listeners. Multiple ambiguous pitches were associated with tone properties such as prominent upper partials and particular autocorrelation profiles. Conclusions. Harmonic tones in a context of contemporary popular music can, in general, convey several ambiguous pitches. The set of perceived pitches depends on both the listener and the listening conditions.
Emmanuel Deruty, David Meredith, Maarten Grachten +5
Sony Computer Science Laboratories, Paris, France · Department of Architecture, Design and Media Technology, Aalborg University, Aalborg, Denmark · Citizen Records, Dijon, France
Much of Western classical music relies on instruments based on acoustic resonance, which produce harmonic or quasi-harmonic sounds. In contrast, since the mid-twentieth century, popular music has increasingly been produced in recording studios, where it is not bound by the constraints of harmonic sounds. In this study, we use modified MPEG-7 features to explore and characterise the evolution of noise and inharmonicity in popular music since 1961. We place this evolution in the context of other broad categories of music, including Western classical piano music, orchestral music, and musique concrète. We introduce new features that distinguish between inharmonicity caused by noise and that resulting from interactions between discrete partials. Our analysis reveals that the history of popular music since 1961 can be divided into three phases. From 1961 to 1972, inharmonicity in popular music, initially only slightly higher than in orchestral music, increased significantly. Between 1972 and 1986, this rise in inharmonicity was accompanied by an increase in noise, but since 1986, both inharmonicity and noise have moderately decreased. In recent years (up to 2020), popular music has remained much more inharmonic than popular music from the 1960s or orchestral music involving acoustic resonance instruments. However, it has become less noisy, with noise levels comparable to those of orchestral music. We relate these trends to the evolution of music production techniques. In particular, the use of multi-tracking may explain the higher inharmonicity in popular music compared to orchestral music. We illustrate these trends with analyses of key artists and tracks.
Emmanuel Deruty, David Meredith, Stefan Lattner
Sony Computer Science Laboratories, 6 rue Amyot, 75005 Paris, France · Department of Architecture, Design and Media Technology, Aalborg University, Rendsburggade 14, 9000 Aalborg, Denmark
Music information retrieval distinguishes between low- and high-level descriptions of music. Current generative AI models rely on text descriptions that are higher level than the controls familiar to studio musicians. Pitch strength, a low-level perceptual parameter of contemporary popular music, may be one feature that could make such AI models more suited to music production. Signal and perceptual analyses suggest that pitch strength (1) varies significantly across and inside songs; (2) contributes to both small- and large-scale structure; (3) contributes to the handling of polyphonic dissonance; and (4) may be a feature of upper harmonics made audible in a perspective of perceptual richness.
Emmanuel Deruty
Sony Computer Science Laboratories, 6 rue Amyot, 75005 Paris, France · Department of Architecture, Design and Media Technology, Aalborg University, Aalborg, Denmark