Non-linear dynamics in mammalian voice production

Abstract

Animal vocalizations range from tonal sounds to irregular atonal sounds and are generated from non-linear oscillations of the vocal folds as well as from turbulent noise in the glottis. Comprehensive study on bioacoustic signals indicates the existence of a diversity of non-linear phenomena, such as limit cycles, subharmonics, biphonation, chaos, and bifurcations in animal vocalizations, which may provide keys to understanding animal communications. In this paper, we review the concept of nonlinear dynamics and its methodology as applicable to bioacoustics. Acoustical analysis of recorded sounds, simulation of a biomechanical model of the voice production system, and physical experiment of the vocal tract and vocal folds are presented to demonstrate non-linear features inherent in animal vocalizations. We focus on source-filter interaction as one of the main regulators of the non-linear property that can lead either to efficient vocalization or to voice instability in animal sounds. A combination of different approaches is suggested to be of great use for extracting the essential features of non-linear dynamics in animal vocalizations.

Subglottal pressure oscillations in anechoic and resonant conditions and their influence on excised larynx phonations

“…Titze et al 11 , Wade et al 12 and Zañartu et al 13 reported on occurrences of sudden pitch frequency jumps and other instabilities when the fundamental frequency of oscillation f o was in the vicinity of the first vocal tract resonance frequency. These phenomena were also observed through numerical simulations by several authors 9 , 14 – 17 . Interactions with subglottal resonances might have a similar influence on the voice source waveform and the vocal fold vibrations as the vocal tract, as observed by Austin et al 18 (using an excised larynx), Zhang et al 19 , 20 (using vocal fold physical models and an excised larynx), or Lucero et al 21 (using vocal fold physical and mathematical models).…”

The perceptual effects of manipulating nonlinear phenomena in synthetic nonverbal vocalizations

“…During regular voiced phonation, the main oscillatorsin most mammals, these are the left and right vocal foldsopen and close in synchrony and at a relatively stable rate, describing a so-called limit cycle in the phase space (Wilden et al 1998;Tokuda 2018). Phonation becomes less stable under certain conditions such as incomplete closure of the vocal folds with high pressure underneath them (subglottal pressure), or when fundamental frequency (f 0 ) approaches or crosses a formanta frequency band amplified by the resonance of the vocal tract (Herzel et al 1995;Wilden et al 1998;Riede et al 2000;Neubauer et al 2004;Cazau et al 2016;Tokuda 2018). Various acoustic irregularities can then occur, particularly if there is an underlying left-right asymmetry in the anatomical structure of oscillators (Herzel et al 1995).…”

“…Various acoustic irregularities can then occur, particularly if there is an underlying left-right asymmetry in the anatomical structure of oscillators (Herzel et al 1995). The exact nomenclature of nonlinearities in acoustic signals varies across disciplines; in bioacoustics, it is common to distinguish pitch jumps, subharmonics, biphonation, and deterministic chaos (Wilden et al 1998;Riede et al 2000Riede et al , 2007Fitch et al 2002;Mann et al 2006;Tyson et al 2007;Blumstein and Recapet 2009;Schneider and Anderson 2011;Tokuda 2018).…”

“…Pitch jumps, or frequency jumps, are abrupt discontinuities of f 0 associated with unstable phonation and described in the vocal repertoire of many animal species (Mann et al 2006;Riede et al 2007;Tyson et al 2007;Schneider and Andersson 2011;Cazau et al 2016). In humans, unwanted frequency jumps are an embarrassment to amateur singers to be avoided when switching between vocal registers (Wilden et al 1998;Tokuda 2018), but they also occur in nonverbal vocalizations (Robb and Saxmann 1988). Overall, however, there are fewer reports and theoretical discussions of pitch jumps than other nonlinearities.…”

“…Chaos is the most challenging nonlinearity to synthesize. Biomechanical two-mass models of vocal folds (Herzel et al 1995;Cazau et al 2016) and even physical models of the vocal tract (Tokuda 2018) have been used to simulate chaotic behaviour and to investigate under what conditions chaos is likely to occur. However, by definition it is difficult to control the exact behaviour of the system once it has switched to a chaotic mode, so these models are less suitable for synthesizing sounds with the desired duration, strength, and dynamics of chaos as well as preserved traces of harmonics with the appropriate f 0 contour.…”