Objective voice analysis of boys with profound hearing loss

SAVE OFFLINE

Objective Voice Analysis of Boys With Profound Hearing Loss *Ali Dehqan and †Ronald C. Scherer, *Zahedan, Iran, yBowling Green, Ohio Summary: Objectives. A critical factor that affects human voice production is hearing because it provides necessary feedback for control of speech. Vocal quality of profoundly hearing-impaired children is often considered deviant from both perceptual and acoustic perspectives. The present study compares selected vocal acoustic parameters of a profound hearing loss group of boys with normal peers. Methods. The subjects were 15 Iranian boys with profound bilateral sensorineural hearing loss and 15 Iranian normal hearing participants matched according to age and sex. The age range of the children with hearing loss was 61–81 months (M ¼ 72.26) (ie, 5.1–6.75 years, M ¼ 6.02 years) and for the normal group the age range was 61–80 months (M ¼ 71.47) (5.08–6.67 years, M ¼ 5.96 years). Each subject phonated 10 /aˆ/ vowels with constant pitch and loudness for maximal phonation times. The mid 3-second portion of each token was analyzed using Dr. Speech 4.3u software (subprogram: Vocal Assessment; Dr. Speech, Tiger Electronics, Seattle, WA). Results. There was a statistically significantly higher fundamental frequency (F0), jitter, and shimmer in the productions of the hearing loss boys compared with the normal hearing boys. Consistent with these findings was a significantly lower value for the harmonics-to-noise ratio measure for the boys with hearing loss. Conclusions. The results of the present study suggest that profoundly deaf children present with greater phonatory instability and spectral noise, with the possible inference of reduced laryngeal control relative to vocal quality. The finding of higher F0 for the boys with profound hearing loss suggests that they use a different control strategy for pitch, an area needing further study. These findings of acoustic and F0 differences of the hearing-impaired boys should be kept in mind for intervention practices especially when the social impact of deafness is considered. Key Words: Hearing loss–F0–Jitter–Shimmer–HNR–Voice analysis–Acoustic analysis–Phonation. INTRODUCTION One of the most important factors that affect the voice during speech is hearing because it provides necessary feedback for control.1 Auditory feedback affects both moment-to-moment and later control of speech.2 Moment-to-moment auditory feedback is important for control of the suprasegmental characteristics of voice and speech, such as fundamental frequency (F0), intensity, and quality. It has been suggested that auditory feedback also influences the control of respiratory, phonatory, and articulatory functions.2–6 Hearing loss is divided into six broad categories on the basis of residual hearing: 0–20 dB is considered average, 21–45 dB reflects a mild hearing loss, 46–60 dB reflects a moderate hearing loss, 61–75 dB reflects a moderately severe hearing loss, 76– 90 dB reflects a severe hearing loss, and greater than 90 dB indicates profound deafness.7 In addition, hearing loss can be divided into three types: conductive hearing loss, sensorineural hearing loss, and mixed hearing loss. Sensorineural hearing loss occurs when there is damage to the cochlea (inner ear) or to the auditory nerve that goes to the brain. This type of hearing loss cannot be medically or surgically corrected and, therefore, the damage is permanent.

Accepted for publication August 31, 2010. From the *Department of Speech therapy, Rehabilitation Faculty, Zahedan University of Medical Sciences, Zahedan, Iran; and the yDepartment of Communication Sciences and Disorders, Bowling Green State University, Bowling Green, Ohio. Address correspondence and reprint requests to Ali Dehqan, Department of Speech Therapy, Rehabilitation Faculty, Zahedan University of Medical Sciences, Ayatollah Kafami, Zahedan, Sistan and Baluchestan, Iran. E-mail: [email protected] Journal of Voice, Vol. 25, No. 2, pp. e61–e65 0892-1997/$36.00 Ó 2011 The Voice Foundation doi:10.1016/j.jvoice.2010.08.006

There is a general agreement in the literature that some of the voice characteristics of deaf people differ considerably from those of speakers with normal hearing.1 Congenitally, deaf speakers tend to have a higher F0 than speakers with normal hearing.8–11 The higher F0 found in the speaking voices of both prepubertal and postpubertal hard of hearing children12 suggest that not hearing the voices of others in their environment leads to using higher pitches than their age peers.13 The voice of profoundly hearing-impaired speakers has been described as monotonous because of the lack of normal pitch variation.9,14 Martony15 measured the F0 of monosyllabic words produced by 22 severely hearing-impaired children. He found that these children showed either excessively large or excessively small variation in both average F0 level and change of F0. Stratton16 studied intonation in sentences produced by 12 severe to profoundly hearing-impaired children and found that they were unable to produce a controlled increase in F0 for stress and that they were unable to sustain a terminal increase in F0 for questions. In English, speakers with profound hearing impairment have been reported to have difficulties in producing either a controlled F0 increase for stress or appropriate terminal F0 changes for questions (F0 increase) and statements (F0 decrease).17 Giusti et al11 evaluated the F0 and its variability in children’s voices with and without sensorineural hearing impairment (severe to profound levels) and compared their voices with normal hearing children’s voices. They found a higher F0 and a larger variability in deaf children compared with normal children. Hearing-impaired children typically present with abnormal vocal quality and laryngeal control.11 For example, Monsen et al18 observed that some deaf people have episodes of diplophonia and voice breaks as well as irregular patterns of F0 and

e62 intensity, suggesting inability to control tension of the vocal folds and subglottal pressure. Difficulties in vocal quality, pitch, and loudness as well as associated perturbation of the glottal waveform have been reported in the hearing loss population.19 Bolfan-Stosic and Simunjak20 measured jitter and shimmer in hearing-impaired children, and results indicated that measures of jitter and shimmer were significantly elevated in the hearing loss group compared with normal controls. The harmonics-to-noise ratio (HNR) characterizes the relationship between two components of the acoustic wave of a sustained vowel: the periodic component and the additional noise coming from the larynx and the vocal tract. The HNR measure was originally introduced as a measure of voice hoarseness.21 Hearing-impaired people may have an unusual voice quality characterized by overaspiration and spectral noise.22 The results of investigations by Thomas-Kersting and Casteel23 showed that hearing-impaired children produced significantly higher spectral noise levels and used more effort during vocalization than hearing children. The motivation for this study was twofold: (1) a comparison of profound hearing-impaired boys with normal hearing boys relative to F0 and vocal perturbation measures may be pertinent to aspects of speech development and rehabilitation, and (2) there are no published data on the acoustic characteristics of Iranian hearing-impaired speakers. The results should provide not only information on the influence of hearing on human voice production but also valuable insights into assessment and planning strategies for speech rehabilitation. METHODS Subjects Subjects consisted of 15 Iranian boys with bilateral profound hearing loss and 15 Iranian normal hearing boys matched according to age. The hearing loss group had auditory thresholds for 500, 1,000, and 2,000 Hz of 95–105 dB for the left ear and 90–105 dB for the right ear.24 The age range of the profound hearing loss boys was 61–81 months (M ¼ 72.26; 5.1–6.75 years, M ¼ 6.02 years) and of the normal group was 61–80 months (M ¼ 71.47; 5.08–6.67 years, M ¼ 5.96 years). No subject of the hearing loss group used hearing aids. All subjects were free of an upper respiratory tract infection for 3 weeks before the test. An otolaryngologist performed a laryngeal examination using indirect laryngoscopy to confirm that no subject had an organic lesion of the vocal folds or other structural deviancy. Furthermore, the normal subjects were perceptually assessed for vocal deviance using the GRBAS scale (which stands for grade, roughness, breathiness, asthenicity, and strain25); a subject was excluded if the rating was higher than zero on any measure. The rating was performed on a voice sample of 1 minute of spontaneous speech. All subjects were also screened for a history of problems with breathing and voice as well as neurological diseases and structural abnormalities of the larynx, mouth, and throat. Subjects who had such problems were excluded from the study. The utterances were performed in a sound-treated room with the subjects seated. The

Journal of Voice, Vol. 25, No. 2, 2011

room noise level was determined by a sound level meter (model: CEL-450, product of Casella CEL; Regent House, Kempston, Bedford, UK) with room noise measured as Min LA: 28.0 dB and Min LC: 40.8 dB. Instrumentation Recordings and signal analyses were carried out using the Dr. Speech 4.3u software program (subprogram: Vocal Assessment; Dr. Speech, Tiger Electronics, Seattle, WA) that was installed on a laptop (Dell Inspiron 6400; Dell Inc, Round Rock, TX; sound card Sigmatel STAC92XX C-Major HD Audio Sigmatel Corp, Austin, TX). The microphone (ECM-717 electret condenser microphone, Sony Corporation, Minato, Tokyo, Japan; frequency response 100–15 000 Hz) was placed on a stand 10 cm from the front of the mouth. The measures acquired from Dr. Speech included the F0, jitter, and shimmer (both based on perturbation quotients26), and HNR. It is noted that the jitter and shimmer values were based on a running average of five points rather than strict cycle-to-cycle differences.26 Voice sample Using tokens of sustained vowels may be preferred over regular speech in vocal acoustic assessment27and thus were used here. After instruction and several test trials, the subjects were instructed to phonate 10 stable /aˆ/ vowels28 continuously for as long as possible, using habitual vocal pitch and loudness and constant quality. They separated the /aˆ/ tokens by 5 seconds. Statistics Data were analyzed with the statistics software SPSS 18.0 for windows (SPSS Corp, Chicago, IL). The Kolmogorov-Smirnov test was used to assess normality of distributions. The statistical differences between the two groups (hearing loss and normal peers) were examined with parametric (independent samples t test) or nonparametric (Mann-Whitney U) tests. RESULTS The mean value and standard deviation for each measure are summarized in Table 1 for the hearing loss boys and normal peers. As illustrated in Figure 1, there was a significant difference between the means of F0 between the two groups. The mean TABLE 1. Comparison Between Normal Hearing Group and Hearing-Impaired Group Normal Parameter F0 Jitter (%) Shimmer (%) HNR

Hearing Impaired

M

SD

M

SD

P-Value

258.18 0.18 1.62 23.53

9.67 0.036 0.31 1.29

396.11 0.74 4.02 13.74

15.21 0.045 0.16 0.77