|Year : 2016 | Volume
| Issue : 2 | Page : 31-35
Comparison of acoustic analysis of voice parameters in children with cochlear implants and normal hearing
A Srividya1, BS Premalatha2, Madhuri Gore1
1 Dr. S. R. Chandrasekhar Institute of Speech and Hearing, Bengaluru, Karnataka, India
2 Department of Speech, Language Studies, Dr. S. R. Chandrasekhar Institute of Speech and Hearing, Bengaluru, Karnataka, India
|Date of Web Publication||13-Oct-2017|
Dr. S. R. Chandrasekhar Institute of Speech and Hearing, Lingarajapuram, Bengaluru - 560 084, Karnataka
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Context: Auditory feedback controls segmental and non-segmental features of speech. The lack of auditory feedback in children with impaired hearing affects voice, speech production and speech perception skills. In children using hearing aids or cochlear implants, better auditory feedback is seen and thus voice and speech skills are improved. The present study looks at the comparison of voice parameters in children with cochlear implants and those with normal hearing. Aims: The study aims to compare voice parameters across a group of male and female children with cochlear implant and normal hearing. Study Design: The design of the study was cross sectional with one time evaluation of the objective voice parameters. Method and Materials: The 15 children with cochlear implant were of the mean age 3.99 years and 1.5 of hearing age for males and mean age of 4.39 years and 1.4 years of hearing age for females. The control group with normal hearing had 30 children with a mean age of 3.81 years. The two groups were evaluated for their voice parameters with MDVP 5105 software of CSL instrument, for sustained vowel /a/. Statistical Analysis: Mean, SD, t-test and significance of variance (p<0.05) were done to compare the voice parameters across both groups of gender. Results: Results showed that significant differences (<0.05) were seen in parameters mean fundamental frequency, perturbation and shimmer among females and including jitter for males when both groups were compared. VTI = Voice turbulence index; SPI = Soft phonation index did not show any significant difference. Conclusion: The study concludes that with early intervention and training, the children with cochlear implants are able to achieve normal voice parameters though the range of variability is higher compared to normal hearing children.
Keywords: Auditory feedback, cochlear implant, multi-dimensional voice profile voice parameters, significant difference
|How to cite this article:|
Srividya A, Premalatha B S, Gore M. Comparison of acoustic analysis of voice parameters in children with cochlear implants and normal hearing. J Laryngol Voice 2016;6:31-5
| Introduction|| |
Auditory feedback controls segmental and suprasegmental features of speech and also moment-to-moment segmental features of speech.,, Hearing loss usually is the cause of improper voice production that can cause social, emotional, and speech limitations, with specific deviation in communication-related to speech and voice. Severe to profound hearing loss affects both speech perception and production skills.
Auditory feedback of voice affects the speech segmental skills and acts as a general controller on articulation, resonance, and respiration. The loss of auditory feedback leads to voice and speech perturbations. Studies provide enough evidence that the voice characteristics of deaf individuals vary considerably from those of individuals with normal hearing (NH).
Congenital deaf speakers tend to have a higher fundamental frequency (f0). Lack of auditory control also affects the control of intensity of voice. Voice of the deaf can be too loud, too soft, or vary irregularly. It is usually described as tense, flat, breathy, harsh with differences in pitch and intonation.
Deviations of voice features may interfere with speech intelligibility and thus affect the social integration of the individual. Children who undergo cochlear implant (CI) as young as 6 months of age, show plasticity in the development of central auditory pathways, reducing the effects of hearing loss, as it facilitates the access to oral speech during the critical periods of language acquisition. Several studies have reported positive effects of early cochlear implantation, which enables hearing and improvement in speech perception.
Justification for the present study
After CI, evaluations usually involve only language assessment. The objectivity of voice parameters and its variations following cochlear implantation has not been included in assessment protocol. Hence, the present study was taken up to emphasize the variations in voice parameters in postcochlear implanted children.
Aim of the study
The present study aims to compare the acoustic analysis of voice in a group of children with CIs with that of age-matched children with NH.
The design of the study was cross-sectional with two groups and one time evaluation.
| Methods|| |
The study comprised two groups of children:
Group 1: CI group
Group 2: NH group.
Group 1: Cochlear implant group
The Group 1 consisted of thirty children of CIs with the age range of 3–5 years. There were 15 males with a mean age of 4.39 years and 15 females with a mean age of 3.99 years [Table 1].
The following inclusion and exclusion criteria were used to recruit children with CIs for the study.
- The degree of hearing loss – should have had severe to profound hearing loss
- Duration of CI usage. The duration of usage of CI should be more than 3 years
- The CI age is the duration of the usage of CI by the child from the day of switch on to the date of sample collection.
- The children should have attended auditory verbal therapy in and around 1 year.
- Children with any syndromic conditions were not considered
- Children who had used hearing aid or CI for <6 months were excluded from the study.
Group 2: Normal hearing group
The NH group consisted of thirty children with the age range of 3–5 years with a mean age of 3.81 years. The group consisted of 15 males and 15 females [Table 1].
The inclusion criteria included the following.
- Should have NH ability
- No history of any speech, language, and hearing disorders
- No history of neurological impairment.
Speech science laboratory of the institute was used for recording purpose. Computer Speech Lab 4500 version 3.4.4 developed by Kay Elemetrics with Multi-Dimensional Voice Profile (MDVP) software 5105 (Version 3.0, KayPentax, New Jersey, USA) was used for both recording as well analysis of stimulus. MDVP software was used as it provides detailed acoustic profile not limited to fundamental frequency and intensity. Instrument was calibrated before recording. Mouth to microphone distance of 10 cm was kept constant for both groups of children.
Vowel /a/ in sustained production was used for recording purpose, for three times. The vowel /a/ was selected as it is easy to produce even by young children and is mainly dependent on acoustic rather than oro-sensitive control with no articulatory obstructions. In addition, a sustained phonation allows MDVP software to detect long-term perturbations in frequency and intensity that would otherwise go undetected with utterances or passages (Campisi, 2005).
The children were made to sit in a comfortable position. They were instructed to produce a sustained vowel /a/ at a most comfortable loudness level. The microphone to mouth distance for the voice recording was kept constant at 10 cm. The child was asked to produce the sustained vowel /a/ as done by the clinician. They were instructed to produce the vowel three times. All the three productions were recorded. The best sample with proper variation in value was taken for analysis.
The following parameters were analyzed using rounded vowel /a/ on MDVP radial diagram. The frequency, intensity, and perturbation parameter were considered for analysis.
The frequency parameters considered were mean fundamental frequency (MF0), average pitch period (T0), the intensity parameter compared was variation in amplitude (vAm), perturbation parameters compared were absolute Jitter (Jita), Jitter percent (Jitt), Shimmer in dB (ShdB), Shimmer percent (Shim), Amplitude Perturbation Quotient (APQ), voice turbulence index (VTI), and soft phonation index (SPI) were determined for every voice sample.
The data were subjected to statistical analysis for mean, standard deviation and Student's t-test for finding out the statistical significance between the two groups using SPSS software. (version 20, 32 bit, Kay Pentax, Lincoln, NewJersey, USA), 32 bit. The level of significance was taken as 0.05.
| Results|| |
Analysis of all the voice parameters was done separately for male and female participants across both the groups.
As seen in [Table 2], the frequency parameters such as mean fundamental frequency (MF0) and average time period (T0) are significantly different statistically for both the groups of children. The intensity parameters such as peak-to-peak amplitude variation (vAm) also show statistical significance across age groups in values. It is seen that unlike in females, only the average pitch period (T0) only shows significant difference statistically though when the mean values were seen to show difference across both the groups for both mean fundamental frequency (MF0) and average pitch period (T0). From [Table 3], when comparing the perturbation parameters across the both groups of children, it is seen that the ShdB and percentage show significant difference along with the APQ. The Jitter and Jitter percentage show no significant difference statistically, but the values are comparably different across the both groups in mean values.
|Table 2: Mean, standard deviation, range of the frequency, and intensity parameters for Group 1 (cochlear implant) and Group 2 (normal hearing) in both genders and value of t-test with significance of difference|
Click here to view
|Table 3: Mean, standard deviation, and range of the perturbation parameters for Group 1 (cochlear implant) and Group 2 (normal hearing) in both gender and value of t-test with significance of difference|
Click here to view
From the analysis [Table 3], it is seen that the significance of difference (P< 0.05) between the two groups of children with respect to P < 0.05. The values of Jitter and jitter in percent showed no significant differences between both the groups (P< 0.05) in females. When the perturbation parameters were compared only ShdB and APQ showed a significant difference statistically.
From the analysis [Table 2] and [Table 3], it may be observed that the comparison between the parameters in males, the values of mean fundamental frequency (MF0), Jitter, Jitter percentage, and Shimmer percent showed no significant differences between both the groups (t< 0.05).
Significant differences were found between both the groups for average pitch period (T0), ShdB, APQ, and peak-to-peak amplitude variation (vAm) in males.
From the analysis [Table 4], it is seen there was statistically significant difference for both VTI and SPI for both genders. The Group 1 (CI) females were seen to have a higher value of SPI compared to the males and the Group 1 (NH).
|Table 4: Mean, standard deviation, range of the voice turbulence index, and soft phonation index parameters for Group 1 (cochlear implant) and Group 2 (normal hearing) in both genders|
Click here to view
| Discussion|| |
The present study observed that the mean fundamental frequency (MF0) of the Group 1 (CI) is lower than the Group 2 (NH) for both genders though it is in the normal range. A significant difference was seen in only females and not in males. Whereas mean average pitch period (T0) was higher in Group 1 (CI) compared to Group 2 (NH) in both genders. Our study is in agreement with the study done by Baudonck et al., 2011 and Miljkovic et al., 2014, Monsen, 1979, who obtained a statistically nonsignificant difference in values of MF0 in both males and females. However, the values of MF0 were higher in range than the NH group in their study.
Coelho et al. found higher maximum and minimum values for MF0 in children with cochlear implantation. Seifert et al. also found lower mean F0 values in children with CIs, similar to our study. However, the participants selected in both the studies were in the higher range and also the age of implantation was later.
Literature suggests that due to poor auditory feedback, the children with hearing loss do not have control on their pitch but children with CIs have a normal F0 range but the range seems to be wider. However, the present study supports the view that earlier intervention and training helps children with hearing loss achieve good auditory feedback and control on their pitch.
The intensity parameters such as variation of amplitude (vAm) mean values were higher in Group 1 (CI) in females and lower in Group 1 (CI) males compared to Group 2. However, statistical significance was seen in both the genders. The study done by Campisi, 2005, on children before cochlear implantation and postcochlear implantation, observed abnormally elevated peak amplitude variation (vAm) indicating the absence of long-term control of amplitude with sustained phonation. They observed after cochlear implantation vAm values returned to normal, thus reestablishing long-term control of amplitude.
Vocal perturbation parameters such as Jitter, Shimmer, and APQ measures indicate the stability of the phonatory system. The present study observed that higher values of perturbation parameters in CI group. This may be due to lack of laryngeal neuromuscular control following poor auditory feedback. The mean frequency perturbation parameters showed that the values are though not statistically significant (except for Jitter percentage in female category) but are higher than normal values. The mean intensity perturbation parameter is observed to be statistically significant both in males and females. The mean value of APQ is lower in Group 1 (CI) compared to Group 2 (NH) in both the genders and also statistically significant. A study done by Bolfan-Stosic and Simunjak  measured Jitter and Shimmer values in hearing impaired children and were significantly elevated compared to normal. However, the range of values of Jitter and Shimmer was within normal values in our study, suggesting positive outcome of implantation and speech and language rehabilitation. The present study in agreement with the study done by Baudonck et al. and Miljkovic et al.
The mean SPI values appear to be higher end in CI group particularly in females compared to NH group, which can be attributed to improper approximation of the vocal folds. It is found to be statistically significant in both genders. The VTI values were statistically significant between two groups in both genders which may be due to incomplete or loose adduction of vocal folds.
| Conclusion|| |
Early intervention in cochlear implantation helps children to achieve acoustically better voice control with adequate auditory feedback and training.
The authors would like to thank children and parents of children who participated in the study and the staff of Dr. S. R. Chandrasekhar Institute of Speech and Hearing.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Fairbanks G. Selected vocal effects of delayed auditory feedback. J Speech Hear Res 1955;20:333-45.
ELman JL. Effects of frequency-shifted feedback on the pitch of vocal productions. J Acoust Soc Am 1981;70:45-50.
Siegel GM, Pick HL Jr. Auditory feedback in the regulation of voice. J Acoust Soc Am 1974;56:1618-24.
Osberger MJ, Mc Garr NS. Speech production characteristics of hearing impaired. In: Speech and Language. New York: Academic Press; 1982. p. 222-83.
Higgins MB, McCleary EA, Carney AE, Schulte L. Longitudinal changes in children's speech and voice physiology after cochlear implantation. Ear Hear 2003;24:48-70.
Leder SB, Spitzer JB, Milner P, Flevaris-Phillips C, Kirchner JC, Richardson F. Voice intensity of prospective cochlear implant candidates and normal hearing adult males. Laryngoscope 1987;97:224-7.
Bauer PW, Sharma A, Martin K, Dorman M. Central auditory development in children with bilateral cochlear implants. Arch Otolaryngol Head Neck Surg 2006;132:1133-6.
Baudonck N, D'haeseleer E, Dhooge I, Van Lierde K. Objective vocal quality in children using cochlear implants: A multiparameter approach. J Voice 2011;25:683-91.
Miljkovic M, Veselinovic M, Sokolovac I, Dankuc D, Komozec Z, Mumovic G. Acoustic analysis of voice in children with cochlear implants. Med Pregl 2014;67(Suppl. 1):32-7.
Monsen RB. Acoustic qualities of phonation in young hearing-impaired children. J Speech Hear Res 1979;22:270-88.
Coelho AC, Bevilacqua MC, Oliveira G, Behlau M. Relationship between voice and speech perception in children with cochlear implant. Pro Fono 2009;21:7-12.
Seifert E, Oswald M, Bruns U, Vischer M, Kompis M, Haeusler R. Changes of voice and articulation in children with cochlear implants. Int J Pediatr Otorhinolaryngol 2002;66:115-23.
Bolfan-Stosic N, Simunjak B. Effects of hearing loss on the voice in children. J Otolaryngol 2007;36:120-3.
[Table 1], [Table 2], [Table 3], [Table 4]