Acoustic and aerodynamic measures in singers : a comparison between genders Medidas acústicas e aerodinâmicas em cantores : comparação entre homens e mulheres

Purpose: Compare acoustic and aerodynamic voice measures between male and female singers. Methods: A cross-sectional, observational, comparative study conducted with a convenience sample. Study participants were 30 male and 30 female singers. Acoustic (vocal intensity and fundamental frequency) and aerodynamic (expiration time, air pressure, expiratory and voice airflow, expiratory volume, aerodynamic power and resistance, acoustic impedance, and aerodynamic efficiency) measures were assessed during emission of the syllable /pá/, at usual frequency and intensity, for seven consecutive times. These emissions enable extraction of air pressure measures (obtained by the plosive consonant /p/, which estimates glottic pressure), as well as of airflow and acoustic voice measures (obtained by the vowel /a/ and the syllable /pá/). Results: Women presented higher values of fundamental frequency compared with those of men. No differences were identified in the evaluation of aerodynamic measures between the groups. Conclusion: Values of aerodynamic measures do not differ between male and female singers.


INTRODUCTION
An individual's vocal quality results from the interaction of aerodynamic forces characterized by expired airflow, myoelastic forces of the larynx, in addition to contributions of the vocal tract filter (1) .The larynx functions as a transducer, converting aerodynamic energy into acoustic energy.The airflow is transformed into acoustic energy by means of the opening and closing cycles of the vocal folds.The respiratory system is responsible for providing the energy needed for phonation; therefore, impairments in this system may affect professionals who use their voice as a working tool (2) .
Voice production presents a multidimensional function, and through assessment of the different vocal parameters is possible to (1) quantify voice quality more accurately; (2) understand the effects of a given treatment, either surgical, pharmacological, or phonotherapeutic; and (3) correlate the different vocal evaluation data for a better understanding of the functional mechanisms involved in phonation (2)(3) .
Assessment of the aerodynamic measures of singers is of utmost importance, because they use respiration as a basis for the adequate production of the quality of their singing voice.Increased airflow and subglottic pressure can occur at any moment, and any respiratory or laryngeal inefficiency may impair vocal production (4) .The specific scientific literature reports that singers differ from the general population regarding the values of aerodynamic measures because of the muscular training they perform (5) .Singers tend to present lower airflow and subglottic pressure values than non-singers, demonstrating that these professionals have greater aerodynamic efficiency (6) .
A study conducted with singers of different musical styles observed that the aerodynamic measures of subglottic pressure and expiratory volume were different, because each singer makes particular vocal adaptations to language, type of music, interpretation, and individual characteristics (4) .
Singing requires control and precision of breathing and phonation and, consequently, greater glottal airflow efficiency (6) .Lack of harmony in one of the functions of these systems can generate imbalances and affect the ability to maintain adequate sound quality (7) .An individual with good respiratory training in singing uses glottal airflow properly.
Aerodynamic analysis of the voice is becoming increasingly viable and common in clinical practice in vocal health as a result of scientific advances.The first aerodynamic measurements were direct, thus invasive, considering that they measured subglottic pressure through puncture of the cricothyroid membrane (8) .Technological advances enabled extraction of aerodynamic measures indirectly, and parameters of airflow and subglottic pressure began to be estimated by intraoral pressure measurements (9) .
Anatomical and functional differences in the larynx and the acoustic characteristics of the voice of men and women have been previously analyzed in the literature comprehensively (10)(11)(12) .However, few studies have investigated differences in the aerodynamic parameters of airflow and subglottic air pressure associated with gender (13)(14)(15) , and no studies analyzing these differences in singers have been found.
Understanding the acoustic and aerodynamic characteristics of the voice in singers and the differences between these aspects regarding gender is of fundamental importance to subsidize the vocal training of singers.It is known that the anatomical differences between the male and female vocal apparatus interfere with the acoustic characteristics of the voice (10)(11)(12) and with the aerodynamic parameters of the spoken voice (13)(14)(15) , but the differences generated by gender in the acoustic and aerodynamic parameters of the voice of singers have not yet been studied in the specific literature.It is assumed that male and female singers present different acoustic and aerodynamic measures of the voice as a result of anatomical differences in the larynx and respiratory system.
In this context, the objective of this study was to compare acoustic and aerodynamic voice measures between male and female singers.

METHODS
This cross-sectional, observational, comparative study was conducted with a convenience sample and approved by the Research Ethics Committee of the Universidade Federal de Minas Gerais (UFMG) under protocol no.CAAE 48085815.2.0000.5149.All participants were informed about the objectives and procedures of the study and signed an Informed Consent Form (ICF).
Study participants were 60 individuals of both genders: 30 men aged 19-48 years (mean=28.6)and 30 women aged 18-36 years (mean=26.1).The groups were matched for age (p=0.208),musical genre, and singing experience time.All participants were selected at singing schools.
Inclusion criteria were as follows: professional or amateur singers with absence of vocal complaints and/or impairments and presence of normal larynx, aged 18-55 years -because this is considered the period of greater vocal stability, thus eliminating any impairment associated with the phase of voice change or with aging.Exclusion criteria comprised smokers, pregnant women, or women in the menstrual period.
Laryngeal assessment was performed through high-speed videolaryngoscopy, which accurately and clearly captures the glottic opening and closing phases, using a Computerized Speech Lab (CSL) Kay Pentax™ Model 6103 console (Lincoln Park, NJ, USA), by a single otolaryngologist.Larynxes whose exams presented complete glottic closure were considered normal, or in the cases of women with posterior triangular cleft, and without lesions in both vocal folds.
All participants were evaluated for presence of vocal and voice quality complaints by a speech-language pathologist with expertise in voice and over five years of practice.Participants without vocal complaint and with neutral vocal quality were included in the study.
With respect to characteristics of the singing voice, the participants were divided into popular and classical singers according to the musical genre reported.In the group of women, 21 (70%) were popular singers with 4-12 years (mean=7.3) of experience and nine (30%) were classical singers with 3-10 years (mean=6.6) of experience.In the group of men, 19 (63.3%) were popular singers and 11 (36.7%) were classical singers, with 5-13 years (mean=7.5)and 4-15 years (mean=6.9) of experience, respectively.
Acoustic and aerodynamic measures of voice were obtained with the CSL Kay Pentax™ Model 6103 program (Lincoln Park, NJ, USA) with a Phonatory Aerodynamic System (PAS) module installed in a Dell  , Optiplex GX260 personal computer and a Direct Sound  professional sound card at the Observatory of Functional Health in Speech-Language Pathology of the Medical School of UFMG (OSF/UFMG).These measures were obtained through emission of the syllable /pá/ at usual frequency and intensity for seven consecutive times.According to the manual "Phonatory aerodynamic system: a clinical manual Kay Pentax™", these emissions enable extraction of air pressure measures (obtained by the plosive consonant /p/, which estimates glottic pressure), as well as of airflow and acoustic voice measures (obtained by the vowel /a/ and the syllable /pá/) (16) .
Aerodynamic measures were recorded using a silicone face mask placed on the participant's mouth.This mask was coupled to a device connected to a pressure transducer, and the intraoral pressure was measured by means of a small-diameter polyethylene catheter inserted into the mask through a lateral orifice and positioned in the central part of the participant's tongue; the other end of the catheter was attached to the pressure transducer (Figure 1).
Acoustic measures of vocal intensity and fundamental frequency (f 0 ) were obtained using a unidirectional condenser microphone attached to the back of the face mask.All signals from the transducer and the microphone were sent to the CSL program for analysis.The acoustic and aerodynamic measurements were conducted in acoustically treated environment, with ambient noise <50 dBNPS defined by an Instrutherm  Model DEC-490 sound pressure level meter.
The following acoustic parameters were analyzed: • Maximum intensity: maximum value of vocal intensity measured in dBNPS.
• Mean intensity: mean value of vocal intensity measured in dBNPS.
• Mean intensity of voiced segments: mean intensity value considering the voiced segments measured in dBNPS.
• Fundamental frequency (f 0 ): mean value of the fundamental frequency measured in Hz.
The aerodynamic measures assessed were grouped into five categories as follows: 1. Temporal Measure: • Expiration time: it measures the expired airflow time, or positive airflow, in seconds.

Air pressure measures:
• Peak air pressure: this measure is the highest value of air pressure observed in the emission of one or more plosive syllables, measured in cm H 2 O.
• Mean peak air pressure: it is mean value of peak air pressure, measured in cm H 2 O.
3 Airflow measures: • Peak expiratory airflow: it is maximum value of the expiratory air flow, measured in L/s.
• Peak voiced airflow: this measure represents the mean airflow of voiced speech segments, measured in L/s.
• Mean airflow during voicing: it is the quotient of the total expiratory air volume by the duration of voiced segments, measured in L/s.

Volume measure
• Expiratory volume: this measure is the total volume of expiratory air measured in liters.

Aerodynamic measures
• Aerodynamic power: is the product of the mean peak air pressure value, the air flow during voicing, and the conversion factor 0.09806, measured in watts.
• Aerodynamic resistance: is the quotient of the mean peak air pressure value by the air flow during voicing, measured in cm H2O/(L/s).It is equivalent to the acoustic impedance measure multiplied by the conversion factor 0.9806.• Acoustic impedance: is the quotient of the mean peak air pressure value by the air flow during voicing measured in dyne s/cm 5 .
• Aerodynamic efficiency: this is a dimensionless value defined in parts per million (ppm) that represents the quotient of acoustic power by aerodynamic power.
Statistical analysis of the data was processed using the Statistical Package for the Social Sciences (SPSS 17.0).First, a descriptive analysis of the data was performed with measures of central tendency and dispersion; after that, the non-parametric Mann-Whitney test was applied to the independent samples for cross analysis of the variables.A 95% confidence interval was considered for all statistical analyses.

RESULTS
Women presented higher values of fundamental frequency (f 0 ) compared with those of men.No differences were identified in the evaluation of aerodynamic measures between genders (Table 1).

Figure 1 .
Figure 1.Illustration of the Kay Pentax  Model 6103 console (Lincoln Park, NJ, USA) in use

Table 1 .
Comparison of measures of habitual vocal emissions between male and female singers Mann-Whitney test.*Statistical significance: p<0.05 Captions: SD = standard deviation