Portuguese Version

Year:  2002  Vol. 68   Ed. 3 - (13º)

Artigo Original

Pages: 378 to 386

PDF PT  

Dancing show singers analysis in pop and opera music styles:perceptual-auditory, acoustic and laryngeal configuration

Author(s): Sueli A. Zampieri(1),
Mara Behlau(2),
Osíris O. C. do Brasil(3)

Keywords: voice quality, voice training, larynx, spectrography

Abstract:
Aim: The aim of this study was to check over laryngeal arrangement used by professional singers of pop music. This idea came from the fact that many times we notice vocal quality changes when we hear pop singers trying to sing opera music. Study design: Clinical prospective. Material and method: A number of 26 subjects was analyzed, 10 males and 16 females, all of them were pop professional singers that worked in dancing shows. A questionnaire was used to check out vocal profile of the subjects. Perceptual-auditory and acoustic analysis was performed. Laryngeal assessment with a flexible endoscope was performed to investigate laryngeal configuration in the pop and opera style. Results: The perceptual-auditory analysis allowed us to observe that the pop singers change their vocal quality when trying to sing a piece of an opera music, increasing vibrato and vocal volume, enhancing vocal resonance and overarticulating the words. The spectrographic analysis didn't show the presence of the singer formant in any of the subject's voices. The laryngeal arrangement of pop singers singing opera music was characterized, in the majority, by an increase of the antero-posterior and median supraglotic closure. Median supraglotic closure cropped up more among the male. Jitter and shimmer values decreased for the sung vowel when compared to the spoken one. These values were statistically significant for the female voices.

Introduction

Popular singers usually start their professional careers relying on their gift to sing, without knowing about singing techniques. It is commonly found that these popular singers have been attracted to singing and music since childhood and suddenly they are discovered by a manager and join the professional life (Oliveira, 1995)1.

Popular singers that work in pubs, night clubs, dancing clubs and parties usually have to sing different musical styles to cater to the taste of the varied public that attend such events (Silva & Campiotto, 1995)2, and in one single night their repertoire may include different musical styles such as samba, country music, pop music, rock, Brazilian popular music and sometimes even pieces of classical music, such as "The Phantom of the Opera", "The Magic Flute" or "Carmine Burana". When we listen to a pop singer singing an opera piece, we realize there are some phonation adjustments that make the vocal quality similar to opera singing, but no one knows for sure what are the real laryngeal adjustments that people use to modify their vocal quality.

Since they are usually cover singers, trying to interpret the hits of famous artists, these popular singers tend to imitate the vocal quality of their idols, frequently developing vocal abuse patterns. Differently from opera singers who usually have a predefined vocal esthetical pattern, popular singers tend to be more appraised by the public if they can come close to the style of famous singers, the so-called superstars, which are defined by Bunch & Chapman (2000)3 as professional singers worldly recognized and constantly targeted by the press.

There are a number of studies that show the differences between popular singing, with phonation adjustments close to speech ones, and opera singing, which requires previous vocal training to adapt vocal loudness and range and whose main characteristics are massive respiratory control, the presence of singers' formant and good vibrato pattern. It is known that in most singing schools the larynx tends to be placed at a stable lower position in the neck, even in high notes, and the changes in tones are a result of lengthening and shortening of the vocal folds. The laryngeal low position in high frequencies, a characteristic of opera singing, as reported by Appleman (1967)4 and studied by Sundberg, Gramming & Lovetri (1993)5 and Miller (1994)6, was also observed by Lovetri, Lesh & Woo (1999)7 during belting singing, which in turn has a very high laryngeal position (Schutte & Miller, 1993)8. These authors believe that different vocal qualities result in different laryngeal adjustments that vary from singer to singer. Similarly, Sundberg (1987)9 reported that individual differences among singers favored different strategies of laryngeal muscles. Campiotto (1998)10 has also noticed that laryngeal position was more related to musical style than opera singing techniques.

The anterior-posterior constriction was observed in different singing styles, such as twang, belting and opera in a study conducted by Yanagisawa, Estill, Kmucha & Leder (1989)11. Hamam, Kyrillos, Bortolai & Figueiredo (1996)12 found that the occurrence of this type of constriction was more frequent in opera singers than in popular singers.

The present study aims at analyzing:

1. the acoustic parameters and laryngeal configuration of popular party singers during spoken production of "é", singing "é", intensity and frequency scales, and singing of a popular song and an opera;

2. laryngeal abnormalities that could be present in the group.

Material and Method

We studied 26 subjects, 10 men and 16 women all professional singers working with popular singing entertainment events with three weekly shows, on average. Ages ranged from 20 to 49 years.

We employed a questionnaire to check the vocal profile of the subjects, performed digital recording of songs sung in both popular and operatic styles to conduct perceptive-auditory and acoustic analysis of the voices, and nasofibroscopic assessment to define laryngeal configuration.

The questionnaire with multiple choice answers addressed time in the profession, number of weekly performances, vocal training, vocal hygiene, complaints and history of vocal pathology. We did not exclude subjects who had history or complaint of laryngeal pathology.

After the questionnaire, we performed digital recording in different areas which were silent but not acoustically treated. We used minidisk brand Sony model MZ-R50 and unidirectional professional microphone brand Shure model SM58. Recordings were made individually encompassing spoken "é", sung "é", first verse of "Canção da América", sung in a popular style and chorus of the song "Con te Partirò", sung in operatic style.

There was no control of the tone of the songs, since the purpose was to check the laryngeal motor gesture and not quality of tuning or musical characteristics.

Subjects underwent nasofibrolaryngoscopy performed by an Otorhinolaryngologist experienced in assessing singing voice, using the nasofiberscope Mashida 3.2mm model ENT-30 III, light source Dyonics Highlight 250W, camera CCD Color Camera Jedmed, videocassette Sony SLV - 72 HF and a 14-inch monitor SONY. We used topical anesthetic tetracaine at 2%, applied to one of the nostrils of the subjects. Patients were asked to produce spoken "é", sung "é", ascending and descending intensity scales with "i", ascending and descending frequency scales with vowel "i", first verse of "Canção da América", and the chorus of the song "Con te Partirò".

We performed acoustic analysis of the digital recording of spoken and sung "é" using the software Dr. Speech Sciences, Tiger Electronics, module Voice Assessment. Based on data analysis, the software provided fundamental frequency (F0) and perturbation values of jitter and shimmer in percentage. We have also performed spectrographic analysis of sung vowel "é" using the module Real Analysis, which is also part of the software Dr. Speech Sciences, to check the presence of singer's formant.

The analyzed parameters were based on the following criteria: fundamental frequency (F0) expressed in Hertz (Hz), jitter PPQ (period perturbation quotient), using an average of 5 points and maximum normal range of 0.5% and shimmer APQ (amplitude perturbation quotient) with 5 point average and maximum normal range of 3%. Singer's formant was analyzed based on the spectrum formed by the simultaneous recording of FFT (Fast Fourier Transform) and LPC (Linear Predictive Coding).

We conducted the perceptive-auditory analysis of the two short songs from the digital recordings in order to confirm whether there had been changes in vocal quality of the subjects during the performance of both styles.

Laryngeal configuration was analyzed using laryngoscopic images. The analysis was visually performed based on the consensus of 30 speech and voice pathologists experienced in voice, considering the following aspects: laryngeal symmetry, presence of glottic chink, type of glottic chink, anterior-posterior and medial supraglottic constriction, vertical laryngeal modification and elongation of vocal folds when changing frequencies, laryngeal tension, laryngeal stability and presence of vibrato.

Laryngeal symmetry was visually assessed using the comparison of right and left sides. The image was considered symmetrical when the hemi-larynges were approximately complementary.

As to glottic closure, it was considered absence of glottic chink when the images presented complete vocal fold coaptation, making the glottic rime disappear completely. The types of chinks considered were triangular chink (posterior, medium-posterior, anterior-posterior), spindle form (anterior, anterior-posterior and posterior), parallel, double, hourglass and irregular.

Anterior-posterior and median constrictions were also visually analyzed; the former was diagnosed when there was a reduction of the vestibule volume caused by the approximation of arytenoid cartilages, reducing or preventing vocal fold visualization. Median constriction was considered as the laryngeal image that presented approximation of lateral vestibular walls, towards the midline, with displacement of the vestibular folds.

When changes in frequency were required, we visually observed whether the laryngeal position in the neck was modified, that is, lower or higher larynx in situations of low and high sounds, or whether the subject used lengthening and shortening of vocal folds during variation of frequency.

In situations of variation in intensity, we visually observed whether louder vocal intensity increased tension, characterized by more glottic coaptation and/or supraglottic involvement.

In the two sung tokens, we observed laryngeal stability by reduction of movements during singing and/or maintenance of the same number of movements employed for the speech.

Vibrato was perceptively and auditorily analyzed and we considered presence of intensity variation.

The findings of laryngeal abnormalities were described by the Otorhinolaryngologist that performed the test.

Acoustic parameters with sung and spoken samples were statistically analyzed using Student's t test for matched data and comparison of gender for all variables. Data presented in Tables 2 and 3 were statistically analyzed using Mann-Whitney test to compare gender and all variables and McNemar test to matched data.

Results

This group consisted of popular party singers who performed with musical groups. The mean number of performances was 3 shows with 5 hours each, and each singer sung on average 13 solo songs in one event. Time in the profession ranged from 5 to 20 years and 19 subjects were full-time singers.

As to singing classes, 20 subjects reported that they had taken classes and 9 of them had studied for at least one year, 7 from one to 5 years, and 4 from six to 10 classes. Only 4 subjects were currently taking classes. Despite that, most of the subjects believed that singing was both a gift and a skill acquired with technique.

When asked about the beginning of their careers, 8 subjects referred that they had started as professionals from the beginning and 16 subjects admitted they had started by imitating famous singers.

As to vocal habits before performances, 14 subjects performed vocal warm-up, 5 had vocal rest, 17 avoided cold drinks, 11 avoided drinks, and only one was concerned about food and 6 of them had no special care with their voices. As to life habits, only 2 singers were smokers and 1 referred alcohol abuse. None of them reported use of illegal drugs.

Table 1 shows the individual and mean values of analyzed acoustic parameters, mean and habitual fundamental frequency (F0), jitter and shimmer during spoken and sung vowel "e".

No singer's formants were detected through spectrographic analysis.

Visual parameters of anterior-posterior and median constriction were considered based on laryngoscopic images during the production of spoken and sung "é", intensity scales and the two singing samples, according to the style of each singer (Table 2).

Table 3 presents the laryngeal and vocal modifications during the production of sung and spoken "é" and the intensity and frequency variations and the two sung samples.

The findings of laryngeal abnormalities observed in the nasofibrolaryngoscopic images are shown in Table 4.

Table 1. Acoustic parameters, mean and habitual fundamental frequency (F0) and (Mode), jitter and shimmer for sustained vowel "é" in sung and spoken production.



Student's T test
Jitter Sung, Male x Female: p=0.068
Jitter Spoken x jitter Sung, Female: p=0.018*
Shimmer Spoken x shimmer Sung, Female: p=0.001*

Table 2. Anterior-posterior and median constriction at emissions of Spoken and Sung "é" in intensity scales and in the two musical samples, according to singing style.



Mann-Whitney Test
Median constriction absent in Spoken E, Male x Female: p=0.047*
Median constriction absent in Sung E, Male x Female: p=0.047*
Median constriction absent in Con te Partirò, Male x Female: p=0.041*
McNemar Test
Anterior-posterior constriction increases, Canção da América x Con te Partirò: p=001*
Median constriction increases, Canção da América x Con te Partirò: p=031*

Table 3. Laryngeal modifications, vibrato analysis and vocal adjustments in production of Sung x Spoken "é", in intensity and frequency scales and in the two musical samples, according to musical style of the singer.



* There were no statistically significant differences in the data.

Table 4. Otorhinolaryngology findings concerning symmetry and laryngeal alterations based on nasofibrolaryngoscopy images and singer's style.

Discussion

To start in the musical career by imitating famous singers is very common among popular singers. When asked about it, 61.53% of the subjects reported that their professional career had started as an imitation. When asked about singing skills, it was interesting to notice that 84.61% believed that singing is both a gift and a skill acquired with training. Despite that fact, it seems that popular singers do not believe that technical coaching is necessary for their profession, since only 4 subjects (15.38%) were currently taking singing classes, showing the poor preparation of this population, similarly to the data collected by Oliveira (1995)1 and Duprat, Eckley, Silva & Costa (1996)13. As to singing classes, 76.92% of the subjects reported they had had some singing classes, but most of them (80%) had classes for less than 5 years. The absence of the appropriate techniques for singing may be detected through the data presented in Table 3. Most of the subjects used vertical variation of the larynx, followed by shortening and lengthening of vocal folds during frequency scales, which suggests their poor technical knowledge.

We found statistically significant differences not only for anterior-posterior constriction but also for median constriction during the operatic singing compared to the popular one. There was also increased laryngeal stability upon the same comparison (Table 3). By correlating these findings with vocal quality modification, improving vocal resonance and overarticulating words during the operatic singing, increasing the vocal volume and vibrato (Table 3), aspects that contribute auditorily to a less tense vocal quality (Sundberg, 1995)14, we assumed that these subjects make intuitively the same laryngeal adjustments necessary to opera singing. This statement differs from what was reported by Miller, Sulter, Schutte & Wolf (1997)15, who advocated that subjects imitate a certain type of vocal production more for its auditory skills than for vocal tract similarity. We also noticed statistically significant difference between genders concerning the presence of median constriction, since it was less frequent in female larynges during sung and spoken "é" and the operatic singing piece.

Vocal quality was auditorily assessed based on the digital recording and not the laryngoscopic assessment, because even though nasofibrolaryngoscopy is considered a good test to assess phonation (Hertegard, 1994)16, which is closer to natural speech and laryngeal biomechanics (Koufman, 1995)17, the situation of the test does not enable great modifications of vocal quality when singing operatic style. Lim, Oates, Phyland & Campbell (1998)18 observed that the use of nasal anesthetic and the presence of the nasofibroscope causes modifications to the vocal quality.

The subjects in the study had difficulty to perform the intensity variation scale. The subjects did not seem to understand the instructions and we had to show them what to do. This issue may be explained by the fact that popular singers do not use intensity variations because they always have amplification equipment, which was also showed by Hamam, Kyrillos, Bortolai & Figueiredo (1996)12. The laryngeal mechanism most commonly used for such a test was increased laryngeal tension. According to Baken (1991)19, changes in voice intensity are a result of increased subglottic pressure and increased glottic resistance to the passage of air.

Acoustic analysis (Table 1) revealed values of fundamental frequency for spoken voice different from the values found in other studies, and the male mean was above the values found by Behlau, Tosi & Pontes (1985)20 and Saviolli (1999)21 and the female mean was below the values found by the former and by Tosi (1998)22. Comparing the sung to the spoken production, we observed that all subjects presented mean fundamental frequency in the sung vowel above that of the spoken vowel, but differentiation between sung and spoken voice was predominantly due to increased in vocal projection, and did not necessarily depend of increased in fundamental frequency. The absence of singer's formant is one more factor that reinforces absence of vocal training in this population, because according to Behlau (1996)23, the detection of singer's formant is one of the clinical contributions of acoustic analysis that enables the differentiation between trained and non-trained voices. As to short-term perturbation rates, the values of jitter and shimmer were reduced in the sung vowel compared to the spoken vowel for most of the subjects. All values of jitter were within the normal range and the values reduced significantly for sung voice as opposed to spoken voice in female subjects. As to shimmer, we observed few rates above the normal range in spoken voice, but all of them were within normal in sung voice. The values of shimmer were statistically significantly reduced in female singing voices. Upon the comparison between genders, the differences between jitter values in singing voice were close to statistical significance (p=0.068), which may be a result of the sample size.

Concerning laryngeal abnormalities (Table 4), there were more male affected larynges than female ones, and the most commonly found pathology in both groups was gastroesophageal reflux. The high number of laryngeal pathologies in subjects that depend on and use their voice professionally may also indicate their poor vocal training, lack of knowledge of laryngeal structures and mechanisms and/or poor vocal hygiene. The same was detected by Teachey, Kahane & Beckford (1991)24 in a study with 30 professional singers with little (2 years) or no formal vocal training, and the most frequent pathological sign was vocal fold nodules. However, even with evidence of poor vocal hygiene, the answers to the questionnaire, similarly to what was reported by Duprat, Eckley, Silva & Costa (1996)13, indicated that popular singers are concerned about vocal health, but they still have not put care into practice.

The data collected indicated the need to perform another study to analyze the imitation of operatic singing by amateur singers in order to observe if the adjustments made by the professional singers studied here are purely physiological or if there is any influence of the singing practice. Koufman, Radomski, Joharji, Russel & Pillsbury (1995)25 observed that amateur singers have high level of laryngeal tension when compared to professional singers, and among the latter, popular singers have more laryngeal tension than operatic singers. We also suggest the study of the perceptive-auditory comparison of operatic singing performed by popular and operatic singers.

Conclusion

1. When singing a part of an opera, popular singers change the vocal quality, increasing vibrato, vocal volume, improving vocal resonance and overarticulating the words;

2. Laryngeal adjustments are different for both popular and operatic singing;

3. There was increased anterior-posterior and median constriction during the operatic singing than during the popular song;

4. Median constriction was more frequent in male than in female subjects in spoken and sung "é" and in operatic singing;

5. The values of jitter and shimmer were lower for sung vowel than for spoken vowel, and the reduction was statistically significant only for female voices;

6. A number of laryngeal abnormalities were detected in the studied group.

REFERENCES

1. Oliveira IB. A educação vocal nos meios de comunicação e arte: a voz cantada. In: Ferreira LP, Oliveira IB, Quinteiro EA & Morato EM. Voz profissional: o profissional da voz. Carapicuíba: Pró-fono; 1995. p.33-43.
2. Silva MAA & Campiotto AR. Atendimento Fonoaudiológico a Cantores Populares. In: Ferreira LP, Oliveira IB, Quinteiro EA & Morato EM. Voz profissional: o profissional da voz. Carapicuíba: Pró-fono; 1995. p.67-90.
3. Bunch M & Chapman J. Taxonomy of Singers Used as Subjects in Scientific Research. J Voice 2000;14: 363-69.
4. Appleman R. The Science of Vocal Pedagogy. Bloomington, Indiana University; 1967. p.87.
5. Sundberg J, Gramming P & Lovetri J. Comparisons of Pharynx, Source, Formant, and Pressure Characteristics in Operatic and Musical Theatre Singing. J Voice 1993;7:301-10.
6. Miller R. The Mechanics of Singing: Coordinating Physiology and Acoustics in Singing. In: Benninger M, Jacobson BH & Johnson AF. Vocal arts medicine: the care and prevention of professional voice disorders. New York: Thieme; 1994. p.61-71.
7. Lovetri J, Lesh S & Woo P. Preliminary Study on the Ability of Trained Singers to Control the Intrinsic and Extrinsic Laryngeal Musculature. J Voice 1999;13:219-26.
8. Schutte HK & Miller DG. Belting and Pop, Nonclassical Approaches to the Female Middle Voice: Some Preliminary Considerations. J Voice 1993;7:142-50.
9. Sundberg J. The science of the singing voice. Illinois, Northern Illinois University; 1987. p.51-91.
10. Campiotto AR. Configurações do trato vocal durante o canto em músicas de três estilos. [Tese de Mestrado em Fonoaudiologia]. Pontifícia Universidade Católica de São Paulo, São Paulo, 1998.
11. Yanagisawa E, Estill J, Kmucha ST & Leder SB. The Contribution of Aryepiglottic Constriction to "Ringing" Voice Quality - A Videolaryngoscopic Study with Acoustic Analysis. J Voice 1989;3:342-50.
12. Haman ACS, Kyrillos LCR, Bortolai AL & Figueiredo VE. Avaliação Vocal de Cantores Líricos e Populares. In: Marchesan IQ, Zorzi JL, Gomes ICD (ed.) Tópicos em Fonoaudiologia, 3. São Paulo: Lovise; 1996. p. 327-39.
13. Duprat AC, Eckley C, Silva MAA & Costa HO. Avaliação Laringológica de Cantores da Noite. In: Marchesan IQ, Zorzi JL, Gomes ICD (ed.) Tópicos em Fonoaudiologia, 3. São Paulo: Lovise; 1996. p. 355-60.
14. Sundberg J. Acoustic and Psychoacoustic Aspects of Vocal Vibrato. In: Dejonckere P, Hirano M & Sundberg J Vibrato. San Diego: Singular; 1995. p.35-62.
15. Miller DG, Sulter AM, Schutte HK & Wolf RF. Comparison of Vocal Tract in Singing and Nonperiodic Phonation. J Voice 1997;11:1-11.
16. Hertegard S. Normal and pathological glottal closure patterns at different ages. British Voice Association. J Voice 1994;3:17-26.
17. Koufman JA. Evaluation of Laryngeal Biomechanics by Fiberoptic Laryngoscopy. In: Rubin JS, Sataloff RT, Korovin GS & Gould WJ (ed.) Diagnosis and Treatment of Voice Disorders. New York: Igaku-Snoin; 1995. p.122-34.
18. Lim VPC, Oates JM, Phyland DJ & Campbell MJ Effects of Laryngeal Endoscopy on the Vocal Performance of Young Adult Females with Normal Voices. J Voice 1998;12:68-77.
19. Baken RJ. An Overview of Laryngeal Function for Voice Production. In: Sataloff RT. (ed.) Professional Voice: the science and art of clinical care. New York: Raven; 1991. p.19-47.
20. Behlau M, Tosi O & Pontes PAL. Determinação da freqüência fundamental e suas variações em altura ("jitter") e intensidade ("shimmer"), para falantes do português brasileiro. ACTA AWHO 1985;4:5-9.
21. Saviolli MRB. Caracterização Vocal dos Alunos da Academia de Polícia Militar do Barro Branco:Sinais e Sintomas Vocais, Proporção s/z e Análise Acústica. [Monografia de Especialização - Centro de Estudos da Voz]. São Paulo; 1999.
22. Tosi DM. Índices de Perturbação da Freqüência Fundamental a Curto e a Longo-Prazo em Mulheres sem Alteração Vocal. [Monografia de Especialização - Centro de Estudos da Voz]. São Paulo; 1998.
23. Behlau M. Considerações sobre a análise acústica em laboratórios computadorizados de voz. In:Araújo R, Pracownik A & Soares L. (ed.) Fonoaudiologia Atual. Rio de Janeiro: Revinter; 1996. p. 93-115.
24. Teachey JC, Kahane JC & Beckford NS. Vocal Mechanics in Untrained Professional Singers. J Voice 1991;5:51-6.
25. Koufman JA, Radomski TA, Joharji GM, Russell GB & Pillsbury DC. Laryngeal Biomechanics of the Singing Voice, 1995. Available at http://www.bgsm.edu/voice. Accessed on 13/11/99.

---

[1] Voice and speech pathologist, Specialist in Voice.
[2] Voice and speech pathologist, Ph.D. in Human Communication Disorders, Federal University of São Paulo - UNIFESP.
[3] Otorhinolaryngologist, Ph.D. in Otorhinolaryngology and Head and Neck Surgery, Federal University of São Paulo - UNIFESP.

Address correspondence to: Sueli A. Zampieri - R. Sebastian Pietro Camandona, 11/ 41 - Bela Vista - Osasco SP - 06080-220 - Tel/Fax: (55 11) 3682.7984 / 9957.3556 - E-mail: suelizampieri@uol.com.br

Conclusion paper submitted to Centro de Estudos da Voz - CEV, Specialization in Voice.

Print: