Harmonics-to-Noise Ratio

Abstract

Distinguishing between vocal changes that occur with normal aging and those that are associated with disease is an important goal of research in voice. Several acoustic measures have been used in an attempt to illuminate the integrity of the vocal mechanism, including harmonics-to-noise ratio (HNR), jitter, and fundamental frequency (F₀). HNR is a measure that quantifies the amount of additive noise in the voice signal; jitter reflects the periodicity of vocal fold vibration. In this study, measures of HNR, jitter and F₀ were used to compare vocal function in three groups of normally speaking women: young adults, middle-aged adults, and elderly adults. Significant differences in HNR emerged between the elderly women and the other two groups. F₀ differences were also apparent between the elderly group and the two younger groups; there were no significant differences in jitter between the three groups. HNR was found to be a more sensitive index of vocal function than jitter. The significant lowering of HNR evident in the elderly speakers may be attributable in part to medications taken by the majority of these elderly subjects.

Introduction

Given the importance of distinguishing between vocal changes that occur with normal aging and those that are associated with disease, finding a robust acoustic measure that reliably signals vocal performance in older adults remains an ongoing challenge. Fundamental frequency (F₀) has been shown to change with aging, increasing with advancing age in elderly men, and decreasing in elderly women.1, 2, 3, 4 By itself, however, F₀ does not yield a sufficiently detailed picture of vocal fold vibratory patterns to differentiate between normal and disordered vocal function. Other acoustic measures such as jitter have yielded inconclusive results in comparisons of younger and older adults' vocal function. For instance, Linville and Fisher⁵ reported higher mean jitter values in elderly women than in younger women. However, jitter levels did not contribute to the overall estimation of the speaker's age. Wilcox and Horii⁶ found significant jitter differences between young and elderly men, while Ringel and Chodzko-Zajko⁷ indicated that jitter values were more related to physiological health than to chronological age. Linville⁸ found considerable overlap in jitter levels among groups of young, middle-aged, and elderly women. Elderly women, on average, did not display more jitter than younger women.

Another acoustic measure that may be a more sensitive index of vocal function is the harmonics-to-noise ratio (HNR). HNR quantifies the relative amount of additive noise in the voice signal.⁹ Additive noise arises from turbulent airflow generated at the glottis during phonation.¹⁰ Inadequate closure of the vocal folds allows excessive airflow through the glottis, giving rise to turbulence. The resulting friction noise is reflected in a higher noise level in the spectrum.¹¹ Noise in the signal may also result from aperiodic vocal fold vibration. The ratio thus reflects the dominance of harmonic (periodic) over noise (aperiodic) levels in the voice, and is quantified in terms of dB. Yumoto, Sasaki, and Okamura¹² reported that adult speakers with normal voice quality obtained HNRs of 7.4 dB and above when producing isolated vowels; other research suggests that values between 11–13 dB are normal for adults.13, 14 Perceptually, HNR reflects voice quality. HNR seems to be one of the parameters that can be used to relate physiological aspects of voice production to a perceptual impression of the voice, because the degree of spectral noise is related to the quality of the vocal output.¹¹ Indeed, it has been reported that HNR is a significant predictor of voice samples rated perceptually as rough.15, 16, 17

A limited number of studies have investigated HNR in elderly speakers. One of the few studies focusing on spectral characteristics in the voices of older adults found that HNR decreased in subjects aged 60 to 99 years old, reflecting a growing voice instability with normal aging.¹⁸ Ramig et al¹⁹ reported that older speakers in poor physical condition showed more spectral noise than older speakers in good phsyical condition or than younger speakers. However, rather than quantifying the degree of noise, Ramig et al used spectrograms to visually determine the level of spectral noise. Mueller¹ made the assumption that because hoarseness is prevalent in the voices of elderly speakers, a lower HNR would be predicted. However, this assumption needs to be supported and quantified in order to distinguish between vocal changes that are within normal limits and those that may be due to pathology.

It is important to establish a normative acoustic database for laryngeal function in older adults. HNR might be a more sensitive index of vocal function in older speakers than is jitter, as the ratio reflects both cycle-to-cycle variability in frequency and amplitude, as well as additive noise generated at the glottis.²⁰ As such, it could be a valuable adjunct in quantifying and differentiating between possible deterioration in voice function in adults who are aging normally and in those with voice problems.

The first purpose of this study, therefore, was to compare HNRs in three age groups of women: younger, middle-aged, and older adults. The second purpose was to compare HNR and jitter measures in terms of sensitivity to potential differences in vocal function in different age groups. The third purpose was to determine F₀ differences among the three speaker groups.