Elsevier

Journal of Voice

Vocal Tremor and Vibrato in the Same Person: Acoustic and Electromyographic Differences

Summary

The goal of this case study was to measure and describe differences between vocal vibrato and essential tremor of the voice in one individual who exhibited both types of modulation. Recordings of spoken and sung vowels produced by the same individual at three effort levels were examined via analysis of acoustic and laryngeal electromyographic (LEMG) signals. Modulation rate, periodicity, and spectral measures of both audio and muscle activation signals revealed generally slower, more prominent, and more regular patterns in sung than in spoken conditions. There was not always a clear correspondence between LEMG and acoustic measures, but both showed differences between the vibrato and tremor of this individual, suggesting differences in the neural bases of these modulations.

Introduction

Pathological vocal tremor and the artistic use of vibrato have long been topics of interest for researchers in the fields of speech science, medicine, and singing. Several authors have considered the similarities and differences in these two vocal behaviors.1, 2, 3 Because of perceptual parallels in the modulations that characterize these phenomena, similar acoustic analysis techniques have been applied to both. Tremor and vibrato introduce oscillations into an otherwise steady voice in the form of frequency (FM) and amplitude modulation (AM). These typically vary in their rate from 4 to 7   Hz. The extent of modulation ranges from barely perceptible to sufficiently severe to cause rhythmic arrests of voicing. Some authors have also examined the periodicity or regularity of vocal modulations.1, 4, 5 These and other accounts have reported that vocal modulation in vibrato—especially FM—is basically sinusoidal, whereas pathological tremor is in many cases less smoothly periodic.

Fundamental to recent theories regarding the origin of these voice modulations is the notion of a mechanism for initiating and sustaining low frequency modulations in laryngeal muscle activity. Both peripheral and central mechanisms have been proposed. In the case of vibrato, it has been hypothesized that a peripheral reflex loop with slowed conduction times and elevated gain may set up an agonist/antagonist resonance in the cricothyroid (CT) and thyroarytenoid (TA) muscles. This leads to rhythmic fluctuations in the tension of the laryngeal muscles.6 The skilled singer can capitalize on these oscillations to produce "a pulsation of pitch, usually accompanied with synchronous pulsations of loudness and timbre, of such extent and rate as to give a pleasing flexibility, tenderness, and richness to the tones."7 Individuals with vocal tremor, on the other hand, may experience shaking as a consequence of a neurological condition such as Parkinson disease or multiple sclerosis. If their tremor is not secondary to another medical diagnosis, it is referred to as essential tremor, which can affect a number of structures, including the vocal tract in 15% of cases. A central neural oscillator is posited as the source of the tremor in all of these neurological conditions. The exact pathophysiology of tremor is not completely understood, in part because of the complexity of connections within the central nervous system. Different diseases may produce tremor by different mechanisms. The clinical manifestations of tremor vary in these conditions. However, metabolic studies in individuals with essential tremor have revealed abnormal cerebellar activation, and functional changes in the olivocerebellar circuit,8, 9 as well as rhythmic neuronal activation in the thalamus.10

A primary difference between the artistic and disordered conditions is that whereas vocal vibrato can be volitionally suppressed for a singer producing "straight tone" vowels, the pathological condition is entirely involuntary. Indeed, whereas vibrato can even be modified in its rate,5 pathological vocal tremor is not responsive to attempts to change it by behavioral means. As a consequence, treatment is mostly limited to the use of laryngeal injection of botulinum toxin (BT) as a means to overcome the most disabling effects of this disorder.11 However, because numerous anatomic structures can contribute to vocal modulation, BT injection into the vocal folds is not effective in reducing vocal tremor that is caused by respiratory or supralaryngeal vocal tract oscillations. Pharmacological agents have met with limited success,12 and neurosurgical intervention, although effective in the regulation of disabling limb tremor,13, 14 is considered too invasive as a treatment for a voice disorder in the absence of limb involvement.

The present study is part of a larger investigation into the mechanisms of vocal tremor, involving laryngeal electromyographic (LEMG) and acoustic recordings. One of the individuals studied had a rapid tremor of the voice, which was present in sustained phonation and conversational speech. The reason for her inclusion in the present report is that she was also a singer accustomed to the use of vibrato, and could produce "sung" and "spoken" vowels during data collection. These two types of phonatory behavior were characterized by modulations that differed perceptually, primarily in their rate. Key differences in the acoustic and muscle activation patterns are presented here to offer insights into the neural control of vocal modulation.

Section snippets

Method

The participant in this investigation was a 23-year-old graduate student at the time of the study. She reported having experienced vocal tremor for about 5 years, and perceptual evaluation revealed a noticeable vocal unsteadiness during speech. When questioned regarding her vocal history, she reported having sung nonprofessionally since her teenage years. When she sang with vibrato, it was observed that her singing voice seemed to be modulated in a different way from her tremulous speaking

Results

Table 1 reports the FM in the acoustic and LEMG signals as a function of loudness level. It also reports COVs in the same signals, and Pearson correlations between them. The values in the table represent the mean of three trials in each condition.

The impression of a difference in modulation between spoken and sung vowels was supported by these acoustic data. The rate of AM and FM in the spoken trials increased from soft to comfortable to loud far more than was the case for sung vowels (Figure 1

Discussion

A number of studies have made reference to the similarities between vocal tremor and vibrato. Although it is true that both are modulations of laryngeal output resulting from changes in muscle tension, the examination of both phenomena in the same individual has allowed valuable insights into some differences between them.

In the Western classical singing tradition, vibrato has generally been attributed to fluctuations in vocal fold tension, particularly as this results from modulation of CT

Acknowledgments

This research was supported by K08-DC00132 from the National Institute of Deafness and Communication Disorders.

References (26)

  • Acoustic analysis of vocal vibrato: a theoretical interpretation of data

    J Voice

    (1989)

  • C. Dromey et al.

    Vibrato rate adjustment

    J Voice

    (2003)

  • L.A. Ramig et al.

    Comparative measures of vocal tremor and vocal vibrato

    J Voice

    (1987)

  • E.H. Buder et al.

    Quantitative and graphic acoustic analysis of phonatory modulations: the modulogram

    J Speech Lang Hear Res

    (2003)

  • W.S. Winholtz et al.

    Vocal tremor analysis with the vocal demodulator

    J Speech Hear Res

    (1992)

  • C. Dromey et al.

    The influence of pitch and loudness changes on the acoustics of vocal tremor

    J Speech Lang Hear Res

    (2002)

  • I.R. Titze et al.

    A reflex resonance model of vocal vibrato

    J Acoust Soc Am

    (2002 May)

  • C.E. Seashore

    The vibrato

    (1932)

  • E.D. Louis

    Behavioral symptoms associated with essential tremor

    Adv Neurol

    (2005)

  • G. Deuschl et al.

    The pathophysiology of tremor

    Muscle Nerve

    (2001 June)

  • S.E. Hua et al.

    Posture-related oscillations in human cerebellar thalamus in essential tremor are enabled by voluntary motor circuits

    J Neurophysiol

    (2005 January)

  • P. Warrick et al.

    Botulinum toxin for essential tremor of the voice with multiple anatomical sites of tremor: a crossover design study of unilateral versus bilateral injection

    Laryngoscope

    (2000)

  • K. Busenbark et al.

    Methazolamide for essential voice tremor

    Neurology

    (1996 November)

  • Cited by (12)

    View full text