Abstract
Objectives.
Report data on acoustic measures of voice in sustained vowels produced by typically developing children, aged 4 to 19 years, to add to the cross-sectional reference values in a pediatric database.
Methods.
Recordings of sustained vowel /ɑ/ phonation were obtained from 158 children (80 males, 78 females) aged 4 to 19 years who were judged to be typically developing with respect to speech and voice. Acoustic analyses were performed with the Multidimensional Voice Program (MDVP™) and the Analysis of Dysphonia in Speech and Voice (ADSV™), both from Pentax Medical.
Results.
Values from both MDVP and ADSV are reported for children in the following age cohorts: 4 to 6 years, 7 to 9 years, 10 to 12 years, 13 to 15 years, and 16 to 19 years.
Conclusion.
The data in this study complement previously published data and contribute to a pediatric reference database useful for research and for clinical practice related to children’s voice. Acoustic parameters most sensitive to age and sex are identified.
1.0. Introduction.
The use of acoustic methods in the study of voice and its disorders is motivated by several factors, but especially by the need for quantitative and unbiased data on phonatory function. Persistent questions have been raised on the validity and reliability of perceptual judgments of voice [1,2], and these questions point to the need for instrumental (“objective”) analyses based on acoustic, aerodynamic, or other methods that can be used to supplement or replace perceptual judgments. Acoustic analysis contributes to the understanding of typical voice development and also has the potential to quantify features of phonation (e.g., harmonics-to-noise ratio, perturbations of vocal frequency or amplitude) that have been used to infer aspects of laryngeal pathophysiology. Most importantly, acoustic methods serve a variety of clinical purposes, such as identifying abnormal voices, contributing to differential diagnosis of voice disorders, evaluating the outcomes of different treatments, and tracking changes associated with a specific therapy or with disease progression. Because acoustic methods are generally noninvasive, they can be used effectively with individuals of different ages and/or with varying states of health. The signal of interest can be recorded in a clinical or research laboratory setting, or remotely by means of readily available technology such as smartphones.
Because voice is highly influenced by the variables of age and sex, these factors are paramount in establishing databases to serve as a reference for the analysis of disordered voice. A primary goal of research on the acoustic properties of normal voice is to construct a reference database for each sex through the lifespan. Progress toward this goal has been summarized in several articles [4,5,6,7], but these emphasize normal voice in adults, and do not present extensive data for children. The reference database for children’s voice has been slowly accumulating, as shown in Table 1, which lists data sources and pertinent descriptions including total number of participants, number of participants of each sex, ages of participants, analysis system and vocal sample, nationality, and comments. The table shows that there is an international interest in using the Multidimensional Voice Program (MDVP™) to analyze children’s voice.
Table 1.
Sources of data on acoustic parameters of children’s voice, showing for each study the total number of participants, number of participants of each sex, ages of participants, analysis system and vocal sample, nationality, and comments.
| Reference | Number of Participants, (Female, Male); NR = not reported | Ages Studied (years) | Analysis System; Vowel Analyzed | Nation | Comments |
|---|---|---|---|---|---|
| Abo-Ras et al. [12] | 100 (58, 42) | 4 to 12 | MDVP; vowel /a/ | Egypt | |
| Akmese et al. [13] | 203 (100, 103) | 4 to 14 | MDVP; vowel /a/ | Turkey | |
| Banik et al. [14]. | 180 (90, 90) | 4 to 12 | MDVP; vowel /a/ | India | |
| Campisi et al. [15] | 100 (90, 90) | 4 to 18 | MDVP; vowel /ɑ/ | Canada | Excluded data for males older than 14 years. |
| Cappellari & Cielo [16] | 23 (NR) | 4 to 6 | MDVP; vowel /a/ | Brazil | |
| Diercks et al. [17] | 40 (male: female ratio of 1.53 to 1.0) | 4 to 17 | MDVP and ADSV; vowel /a/ (also included four sentences, and the first sentence of the “rainbow passage”) | USA | Data for females and males not reported separately. |
| Hill et al. [18] | 50 (male: female ratio of 1.08 to 1) | 4 to 17 | MDVP; vowel /a/ | USA | |
| Jotz et al. [19] | 50 (all males) | 3 to 10 | MDVP; vowel /a/ | Brazil | Data for boys older than 5 years included in this summary |
| Maturo et al. [20]. | 335 (165, 170) | 4 to 18 | MDVP; vowel “ah” | USA | |
| Okuda et al. [21] | 88 (41, 47) | 6 to 9 | MDVP; vowel /ɑ/ | Japan | Data reported only for PPQ, APQ, and NHR; means estimated from graphs in article |
| Tavares et al. [22] | N = 240 (120, 120) | 4 to 12 | MDVP; vowel /a/ | Brazil | |
| Toki et al. [23] | 523 (296, 227) | 4 to 17+ years | MDVP; vowels /a/ and /i/ | Greece |
Acoustic analysis of voice can be accomplished efficiently using specialized software or hardware/software systems. Several such systems are available. This report pertains to one that is frequently mentioned in the literature, the Computerized Speech Lab software program CSL 4500 with software modules Multidimensional Voice Program (MDVP™) and Analysis of Dysphonia in Speech and Voice (ADSV™), from PENTAX Medical. The software employs algorithms that can run semi-automatically to generate quantitative data on voice using high quality recordings with a high sampling rate. MDVP is most suited to analysis of sustained phonation and requires analysis over an interval of at least 3 s to ensure stability of measurements. ADSV™ uses spectral- and cepstral-based analyses that can be performed on different types of speech samples. One parameter sensitive to disordered voice, the Cepstral/Spectral Index of Dysphonia (CSID), is a multivariate estimate of dysphonia severity based on the CAPE-V perceptual assessment scale [8). Validation of this measure has been reported [9,10].
The purpose of this report is to enrich the database on acoustic measures of voice in typically developing children of both sexes by reporting cross-sectional reference data on several variables from both MDVP and ADSV for ages 4 to 19 years, and to compare the results with previously published data as a step toward compilation of an international database on pediatric voice. An additional objective is to determine if MDVP and ADSV values change systematically with age in each sex, which is a fundamental step in evaluating the adequacy of a database and identifying maturational effects on voice.
2.0. Methods
2.1. Participants.
Participants were 158 children (80 males, 78 females) aged 4 to 19 years who were considered to be typically developing with respect to speech and voice, as judged by two speech-language pathologists who listened to the children’s productions of sustained vowels, isolated words and short sentences. The number of female (F) and male (M) participants in each age group was as follows: 4–6 years (13F; 15M), 7–9 years (15F; 15M), 10–12 years (15F; 15M), 13–15 years (15F; 15M), and 16–19 years (20F; 20M). These 2- or 3-year groupings were used to provide the optimum age-group comparisons with previously published studies and were considered to be sufficient to establish developmental patterns for the prepubertal (7 years and younger), peripubertal (8 to 10 years), pubertal (11 to 16 years) and postpubertal (older than 16 years) stages. The youngest age Group 1 can be regarded as prepubertal, Group 2 as peripubertal, Groups 3 and 4 as pubertal, and Group 5 as postpubertal, using approximately the same categories as in [11]. The participants in the present study were part of a larger project to obtain normative data on acoustic properties of speech. Data on formant frequencies and fundamental frequency (Fo) of the corner vowels produced in words by many of these speakers were reported previously [11]. Therefore, the present study on sustained phonation of vowel /ɑ/ complements the earlier report to give a developmental profile of acoustic measures of voice and vowel articulation.
2.2. Materials.
The vowel recordings were obtained as part of a larger study that also included isolated words and short sentences. Participants were seated in a quiet room and instructed to perform the various tasks of the research protocol, one part of which was sustained production of vowel /ɑ/. Voice signals were recorded by a unidirectional Shure SM48 microphone (Shure Inc.) that was placed at an angle of 45 degrees approximately 15 cm from the speaker’s mouth and was stabilized using a floor stand. The microphone was directly connected to a Marantz PMD660 (Marantz Professional) digital audio recorder that digitized the voice signals at 48 kHz with 16-bit resolution. The targeted recording level on the recorder’s volume unit (VU) meter was between 6 to 12 dB below the maximum level. Before the vowel sample was recorded, the participants counted from one to five and/or repeated two practice stimuli while recording level adjustments were made to ensure good signal quality.
2.3. Measurements.
Acoustic analyses were performed with the Multidimensional Voice Program (MDVP) (model 5101) and Analysis of Dysphonia in Speech and Voice (ADSV) (model 5109) software options for the Computerized Speech Laboratory (model 4500; KayPENTAX). The parameters analyzed by the two programs are listed in Appendix 1. MDVP calculates up to 33 parameters of voice, and ADSV calculates a smaller number of parameters CPP and associated statistics, L/H Spectral Ratio and associated statistics, Fo, and CSID (sex appropriate). The present study provides data on parameters frequently reported in the literature. The vowel recordings to be analyzed were inspected in a waveform display and a segment of 3 s duration was selected for analysis, excluding the initial and final portions of the waveform so as to avoid onset and offset features of phonation (e.g. hard glottal attack, breathy or strained final phase of phonation). The ADSV analysis was performed on the same 3 s interval used for the MDVP analysis. Data for females and males were visually examined in group scatterplots for each MDVP and ADSV variable. Outliers were identified with a criterion of 3 standard deviations from the group mean, which excluded highly deviant values without constraining natural variability among voices. With this criterion, less than 1% of the data was excluded from analysis and these pertained largely to three children, two girls and one boy.
2.4. Research ethics.
This research was approved by the Health Sciences Institutional Review Board of the University of Wisconsin-Madison.
3.0. Results
Statistical summaries of 11 MDVP and 5 ADSV parameters obtained in this study are reported in Table 2 and 3 for females and males aged 4 to 19 years. Table 2 table shows for each parameter and each sex group across the age range studied the mean and standard deviation, median, and range. Individual variability was large for most parameters, as indicated by the values of standard deviation and range. Table 3 shows the cross-sectional statistics (means and standard deviations) for the 5 age groups. The parameters Fo, vFo, Jitt, PPQ, L/H standard deviation, and sex-appropriate CSID generally show decreasing values with age for both females and males. The parameters CPP standard deviation and L/H ratio show generally increasing values across age for both sexes. For the other parameters, the patterns across age vary by sex, with females exhibiting a more consistent pattern of decreasing values with age (as seen for all parameters except vAm, SPI, CPP, and the aforementioned CPP standard deviation and L/H ratio). There is considerable inter-individual variability for most parameters.
Table 2.
Shown for selected MDVP and ADSV parameters are the means, standard deviations (s.d.), medians, and ranges for the female and male groups. Fo statistics except for range are not shown for males because of the large and nonlinear changes in Fo beginning by about 10 to 12 years of age [11,20].
| Parameter/Measure (unit) | Mean (s.d.) | Median | Range (minimum to maximum) | |
|---|---|---|---|---|
| Fo (Hz) | 235.3 (30.79) NA |
239.4 NA |
116 – 287 78.5 – 290.6 |
|
| STD (Hz) | 4.89 (3.03) 4.51 (6.30) |
4.36 3.41 |
1.337 – 15.942 0.565 – 18.233 |
|
| vFo (%) | 2.05 (1.93) 1.92 (1.15) |
1.70 1.58 |
0.61 – 6.33 0.54 – 6.404 |
|
| Jitt (%) | 1.87 (1.36) 1.36 (1.0) |
1.42 1.08 |
0.297 – 7.466 0.271 – 4.91 |
|
| PPQ (%) | 1.09 (0.76) 0.84 (0.71) |
0.83 0.64 |
0.173 – 3.406 0.161 – 4.225 |
|
| vAm (%) | 18.56 (8.06) 20.23 (8.97) |
16.11 19.16 |
6.94 – 42.57 5.961 – 44.482 |
|
| Shim (%) | 5.55 (2.17) 5.23 (2.42) |
5.03 4.63 |
2.42 – 13.6 1.352 – 13.091 |
|
| APQ (%) | 3.872 (1.46) 3.80 (1.58) |
3.53 3.53 |
1.676 – 9.736 1.14 – 9.45 |
|
| NHR | 0.143 (0.04) 0.153 (0.06) |
0.13 0.14 |
0.092 – 0.326 0.089 – 0.412 |
|
| VTI | 0.062 (0.023) 0.06 (0.022) |
0.058 0.055 |
0.029 – 0.18 0.022 – 0.152 |
|
| SPI | 7.20 (4.87) 8.25 (5.57) |
6.484 6.534 |
0.726 – 19.348 1.237 – 30.21 |
|
| CPP(dB) | 10.05 (1.99) 11.67 (2.56) |
9.960 11.534 |
2.858 – 13.49 6.233 – 17.711 |
|
| CPP standard deviation (dB) | 0.92 (0.42) 1.167 (0.50) |
0.892 1.039 |
0.325 – 2.648 0.396 – 2.878 |
|
| L/H ratio (dB) | 28.19 (3.53) 30.39 (4.52) |
28.660 30.534 |
18.785 – 36.106 17.526 – 39.628 |
|
| L/H ratio standard deviation (dB) | 1.79 (0.61) 2.10 (0.72) |
1.60 1.98 |
1.0 – 4.26 1.02 – 4.81 |
|
| CSID sex appropriate | 23.19 (11.85) 29.26 (16.50) |
22.304 30.692 |
3.039 – 52.229 −7.993 – 74.756 |
|
Table 3.
Means (and standard deviations) for MDVP and CSID parameters for females (F) and males (M) in each of the 5 age groups. CPP SD is the standard deviation of CPP, and L/H ratio SD is the standard deviation of the L/H ratio.
| Parameter (measure unit) | Group 1 4–6;11 yrs | Group 2 7–9;11 yrs | Group 3 10–12;11 yrs | Group 4 13–15;11 yrs | Group 5 16–19;11 yrs | |
|---|---|---|---|---|---|---|
| Fo (Hz) | 257.0 (15.0) 245.2 (25.48) |
244.8 (22.9) 241.6 (31.08) |
253.9 (24.8) 239.4 (29.33) |
219.2 (32.2) 151.4 (43.33) |
214.5 (27.7) 107.3 (20.33) |
|
| STD (Hz) | 7.32 (3.72) 7.52 (4.16) |
6.65 (3.79) 4.30 (1.42) |
4.84 (1.97) 7.68 (1.62) |
3.64 (1.75) 2.61 (1.73) |
3.25 (1.81) 1.41 (0.75) |
|
| vFo (%) | 2.79 (1.55) 3.09 (1.64) |
2.72 (1.54) 1.75 (0.52) |
1.90 (0.73) 1.81 (0.64) |
1.67 (0.74) 1.76 (1.04) |
1.49 (0.74) 1.36 (0.82) |
|
| Jitt (%) | 2.53 (3.72) 2.17 (1.36) |
2.35 (1.69) 1.14 (0.48) |
1.84 (0.91) 1.89 (0.78) |
1.54 (0.91) 1.20 (0.72) |
1.38 (0.86) 0.90 (0.94) |
|
| vAm (%) | 17.50 (5.41) 20.61 (6.17) |
19.76 (9.58) 24.16 (7.69) |
23.91 (9.72) 19.52 (6.16) |
15.35 (5.04) 16.13 (8.38) |
16.97 (7.42) 20.71 (12.41) |
|
| Shim (%) | 6.67 (2.58) 6.86 (3.12) |
6.81 (2.46) 4.48 (1.43) |
5.58 (2.11) 5.00 (1.58) |
4.92 (1.84) 4.78 (1.84) |
4.39 (0.98) 6.05 (2.43) |
|
| PPQ (%) | 1.42 (1.05) 1.15 (0.86) |
1.39 (1.02) 0.67 (0.28) |
1.06 (0.52) 1.12 (0.45) |
0.91 (0.54) 0.68 (0.37) |
0.81 (0.54) 0.53 (0.56) |
|
| APQ (%) | 4.60 (1.87) 4.78 (1.86) |
4.72 (1.68) 3.19 (0.98) |
3.78 (1.33) 3.46 (0.90) |
3.56 (1.84) 4.59 (1.12) |
3.10 (0.64) 4.59 (1.86) |
|
| NHR | 0.16 (0.06) 0.17 (0.06) |
0.16 (0.04) 0.13 (0.02) |
0.14 (0.02) 0.13 (0.02) |
0.14 (0.05) 0.16 (0.04) |
0.13 (0.02) 0.16 (0.06) |
|
| VTI | 0.08 (0.04) 0.08 (0.03) |
0.07 (0.02) 0.05 (0.01) |
0.06 (0.02) 0.05 (0.01) |
0.05 (0.01) 0.05 (0.01) |
0.06 (0.01) 0.06 (0.02) |
|
| SPI | 4.49 (3.32) 4.75 (3.04) |
5.36 (3.13) 5.23 (4.79) |
8.91 (7.77) 8.98 (5.46) |
8.97 (3.70) 11.39 (6.91) |
7.69 (4.08) 9.83 (4.92) |
|
| CPP (dB) | 9.28 (1.79) 9.70 (1.79) |
9.46 (1.29) 11.02 (1.72) |
9.175 (1.84) 9.82 (1.82) |
10.82 (1.81) 12.67 (1.96) |
11.06 (1.54) 14.08 (2.14) |
|
| CPP SD (dB) | 1.02 (0.38) 1.04 (0.38) |
0.98 (0.24) 0.94 (0.25) |
1.16 (0.47) 0.99 (0.24) |
0.79 (0.56) 1.50 (0.63) |
0.75 (0.29) 1.30 (0.60) |
|
| L/H ratio (dB) | 25.38 (4.86) 26.37 (4.86) |
28.96 (2.89) 30.36 (4.61) |
28.16 (4.50) 28.94 (2.76) |
29.24 (2.62) 31.81 (3.43) |
28.67 (3.25) 33.04 (4.02) |
|
| L/H ratio SD (dB) | 2.06 (0.54) 2.24 (0.54) |
2.04 (0.80) 2.12 (0.75) |
1.76 (0.31) 2.16 (0.93) |
1.75 (0.55) 1.96 (0.69) |
1.50 (0.31) 2.03 (0.68) |
|
| CSID | 32.51 (14.74) 42.72 (14.74) |
27.43 (6.95) 29.87 (10.20) |
29.30 (11.27) 37.58 (10.86) |
16.88 (10.56) 24.85 (13.61) |
14.09 (8.46) 16.86 (18.18) |
|
Direct comparisons of data from the present study and previous studies should be done with the recognition that they differ in several respects (e.g., recording equipment and environment) that could influence the results. With this caveat in mind, comparisons of the present results with data from previous studies using MDVP and/or ADSV are given in Table 4 (females) and Table 5 (males). The data are derived from samples of sustained production of the open vowels /ɑ/ or /a/. The results are given in age groups that were judged to be most appropriate to summarize published studies given that some studies did not specify values in yearly increments but rather in age groups. As noted in section 2.1, the groups can be considered to be in approximately the same pubertal groups as in earlier studies [11,24], that is, prepubertal (7 years and younger), peripubertal (8 to 10 years), pubertal (11 to 16 years) and postpubertal (older than 16 years). Data from the present study are denoted by an asterisk (*) and bold font.
Table 4.
Cross-sectional MDVP and ADSV data for females compiled from various sources. Data from the present study are denoted by asterisk (*) and bold font (individual age-groups’ values same as Table3, and across age values same as Table 2). Note: data for females and males are combined in two studies [15,17].
| Parameter (measure unit) | Group 1 4–6;11 yrs | Group 2 7–9;11 yrs | Group 3 10–12;11 yrs | Group 4 13–15;11 yrs | Group 5 16–19;11 yrs | Across Ages 4 to 19 yrs | |
|---|---|---|---|---|---|---|---|
| Fo (Hz) |
* 257.0 [12] 289.59 [13] 299.33 [14] 261.94 - [16] 252.81 - [20] 272.3 [22] 257.14 [23] 281.3 |
* 244.8 [12] 279.39 [13] 287.6 [14] 267.77 - - - [20] 254.7 [22] 230.48 [23] 271.7 |
* 253.9 [12] 253.17 [13] 266.0 [14] 252.97 - - - [20] 254.0 [22] 234.09 [23] 262.1 |
* 219.2 - [13] 240.3 - - - - [20] 219.0 - [23] 231.2 |
* 214.5 - - - - - - [20] 224.7 - [23] 239.0 |
* 235.3 - - - [15] 279.05 - [18] 244.57 [20] 247 - [23] 263.4 |
|
| STD (Hz) | * 7.32 | * 6.65 | * 4.84 | * 3.64 | * 3.25 | * 4.89 | |
| vFo (%) |
* 2.79 [13] 4.03 - [16] 4.17 |
* 2.72 [13] 3.40 - - |
* 1.90 [13] 2.57 - - |
* 1.67 [13] 1.74 - - |
* 1.49 - - - |
* 2.05 - [15] 1.75 - |
|
| Jitt (%) |
* 2.53 [12] 1.62 [13] 1.51 [14] 0.54 - - - [20] 1.44 [22] 1.64 [23] 1.313 |
* 2.35 [12] 1.54 [13] 1.24 [14] 0.38 - - - [20] 1.43 [22] 1.621 [23] 1.50 |
* 1.84 [12] 1.69 [13] 1.2 [14] 0.53 - - - [20] 0.94 [22] 1.67 [23] 1.47 |
* 1.54 - [13] 1.18 - - - - [20] 1.16 - [23] 2.67 |
* 1.38 - - - - - - [20] 1.16 - [23] 1.78 |
* 1.87 - - - [15] 1.24 [17] 1.68 [18] 1.57 [20] 1.24 - [23] 1.60 |
|
| vAm (%) |
* 17.5 [13] 35.20 - [16] 28.88 |
*19.76 [13] 28.26 - - |
* 23.91 [13] 23.17 - - |
* 15.35 [13] 29.67 - - |
* 16.97 - - - |
* 18.56 - [15] 15.10 - |
|
| Shim (%) |
* 6.67 [12] 3.5 [13] 5.59 [14] 0.16 - - - [20] 3.09 [22] 5.048 [23] 5.83 |
* 6.81 [12] 4.98 [13] 4.83 [14] 0.15 - - - [20] 3.04 [22] 4.74 [23] 5.82 |
* 5.58 [12] 3.75 [13] 4.13 [14] 0.16 - - - [20] 2.57 [22] 4.22 [23] 5.86 |
* 4.92 - [13] 4.06 - - - - [20] 2.62 - [23] 6.68 |
* 4.39 - - - - - - [20] 2.58 - [23] 6.15 |
* 5.50 - - - [15] 3.35 [17] 3.48 [18] 3.38 [20] 2.81 - [23] 5.95 |
|
| PPQ (%) |
* 1.42 [13] 0.91 - [16] 1.02 [21] 0.6 [22] 0.955 |
* 1.392 [13] 0.71 - - [21] 0.38 [22] 0.959 |
* 1.06 [13] 0.73 - - [21] 0.7 [22] 1.015 |
* 0.912 [13] 0.715 - - - - |
* 0.806 - - - - - |
* 1.09 - [15] 0.71 - - - |
|
| APQ (%) |
* 4.604 [13] 3.97 - [16] 6.4 [21] 2.75 [22] 3.543 |
* 4.715 [13] 3.56 - - [21] 2.5 [22] 3.289 |
*3.776 [13] 3.10 - - [21] 2.95 [22] 2.858 |
* 3.556 [13] 2.935 - - - |
* 3.10 - - - - |
* 3.87 - [15] 2.32 - - |
|
| NHR |
* 0.16 [12] 0.13 [13] 0.15 [14] 0.11 - [16] 0.2 - - [20] 0.12 [21] 0.12 [22] 0.14 [23] 0.16 |
* 0.16 [12] 0.13 [13] 0.14 [14] 0.10 - - - - [20] 0.12 [21] 0.10 [22] 0.14 [23] 0.16 |
* 0.14 [12] 0.13 [13] 0.14 [14] 0.10 - - - - [20] 0.11 [21] 0.12 [22] 0.13 [23] 0.16 |
* 0.142 - [13] 0.125 - - - - - [20] 0.12 - - [23] 0.22 |
* 0.13 - [13] 0.125 - - - - - [20] 0.09 - - [23] 0.16 |
* 0.14 - - - [15] 0.11 - [17] 0.12 [18] 0.11 [20] 0.11 - - [23] 0.16 |
|
| VTI |
* 0.08 [13] 0.05 - [16] 0.06 |
* 0.07 [13] 0.04 - - |
* 0.06 [13] 0.04 - - |
* 0.05 [13] 0.035 - - |
* 0.06 - - - |
* 0.062 - [15] 0.05 - |
|
| SPI |
* 4.49 [13] 5.16 - [16] 3.64 [22] 5.02 |
* 5.36 [13] 8.87 - - [22] 6.01 |
* 8.91 [13] 10.94 - - [22] 6.69 |
* 8.97 [13] 16.34 - - - |
* 7.69 - - - - |
*7.20 - [15] 9.80 - - |
|
| CPP (dB) |
* 9.28 - |
* 9.46 - |
* 9.175 - |
* 10.82 - |
* 11.06 - |
* 10.05 [17] 9.82 |
|
| CPP standard Deviation (dB) | * 1.02 - |
* 0.98 - |
* 1.162 - |
* 0.790 - |
* 0.746 - |
* 0.925 [17] 0.96 |
|
| L/H ratio (dB) | * 25.38 | * 28.96 | * 28.16 | * 29.24 | * 28.67 |
*28.19 [17] 36.47 |
|
| L/H ratio standard deviation (dB) |
* 2.06 - |
* 2.04 - |
* 1.76 - |
* 1.75 - |
* 1.50 - |
* 1.79 [17] 1.77 |
|
| CSID |
* 32.51 - |
* 27.43 - |
* 29.30 - |
* 16.88 - |
* 14.09 - |
* 23.19 [17] 21.84 |
|
Table 5.
Cross-sectional MDVP and ADSV data for males compiled from various sources. Data from the present study are denoted by asterisk (*) and bold font (individual age-groups’ values same as Table3, and across age values same as Table 2). Note: data for females and males are combined in two studies [15,17].
| Group 1 4–6;11 yrs | Group 2 7–9;11 yrs | Group 3 10–12;11 yrs | Group 4 13–15;11 yrs | Group 5 16–19;11 yrs | Across Ages 4 to 19 yrs | |
|---|---|---|---|---|---|---|
| Fo (Hz) |
* 245.2 [12] 276.7 [13] 292.2 [14] 266.5 - [16] 252.81 [19] 243.1 [20] 271.3 [22] 275.1 [23] 287.35 |
* 241.55 [12] 244.9 [13] 266.2 [14] 253.9 - - [19] 228.1 [20] 247 [22] 227.3 [23] 271.86 |
* 239.36 [12] 230.9 [13] 250.6 [14] 237.7 - - [19] 219.4 [20] 222.3 [22] 222.5 [23] 256.11 |
* 151.40 - [13] 186.7 - - - - [20] 162 - [23] 195.32 |
* 107.29 - - - - - - [20] 109 - [23] 145.43 |
* 191.4 - - - [15] 279.0 - - [20] 202.3 - [23] 251.5 |
| STD (Hz) | * 7.52 | * 4.30 | * 7.68 | * 2.61 | * 1.41 | * 3.89 |
| vFo (Hz) |
* 3.09 [13] 5.26 - [16] 4.17 |
* 1.75 [13] 4.18 - - |
* 1.81 [13] 2.55 - - |
* 1.76 [13] 2.04 - - |
* 1.36 - - - |
*1.92 - [15] 1.75 - |
| Jitt (%) |
* 2.17 [12] 2.18 [13] 1.53 [14] 1.14 - [19] 2.14 [20] 1.94 [22] 1.71 [23] 1.14 |
* 1.140 [12] 2.04 [13] 1.22 [14] 1.11 - [19] 1.8 [20] 1.62 [22] 1.53 [23] 1.78 |
*1.89 [12] 1.66 [13] 1.23 [14] 0.68 - [19] 1.21 [20] 1.43 [22] 1.71 [23] 1.42 |
* 1.20 - [13] 1.16 - - - [20] 1.35 - [23] 1.52 |
* 0.90 - - - - - [20] 1.06 - [23] 1.89 |
*1.36 - - - [15] 1.24 - [20] 1.51 - [23] 1.51 |
| vAm (%) |
* 20.61 [13] 33.20 [16] 28.88 |
* 24.16 [13] 27.58 - |
* 19.52 [13] 24.03 - |
* 16.13 [13] 25.46 - |
* 20.71 - - |
* 20.23 - [15] 15.10 |
| Shim (%) |
* 6.860 [12] 4.09 [13] 5.31 [14] 0.52 - [19] 6.66 [20] 3.65 [22] 4.37 [23] 5.25 |
* 4.477 [12] 5.34 [13] 4.62 [14] 0.45 - [19] 6.56 [20] 2.88 [22] 4.24 [23] 6.34 |
* 5.0 [12] 2.95 [13] 4.41 [14] 2.08 - [19] 5.34 [20] 2.76 [22] 4.01 [23] 5.82 |
* 4.78 - [13] 3.81 - - - [20] 3.35 - [23] 5.09 |
* 6.05 - - - - - [20] 3.11 - [23] 5.82 |
* 5.48 - - - [15] 3.35 - [20] 3.21 - [23] 5.80 |
| PPQ % |
* 1.154 [13] 0.927 - [16] 1.02 [19] 1.26 [22] 0.85 |
* 0.671 [13] 0.737 - - [19] 1.08 [21] 0.872 |
* 1.116 [13] 0.737 - - [19] 0.735 [22] 1.024 |
* 0.681 [13] 0.70 - - - - |
* 0.530 - - - - - |
* 0.837 - [15] 0.71 - - - |
| APQ % |
* 4.78 [13] 4.0 - [16] 6.4 [19] 4.61 [21] 2.25 [22] 3.133 |
* 3.19 [13] 3.38 - [19] 4.76 [21] 2.1 [22] 3.255 |
* 3.46 [13] 3.38 - [19] 3.78 [21] 2.1 [22] 2.789 |
* 4.59 [13] 2.84 - - - - - |
* 4.59 - - - - - - |
* 3.95 - [15] 2.32 - - - - |
| NHR |
* 0.168 [12] 0.13 [13] 0.153 [14] 0.13 - [16] 0.2 [20] 0.123 [21] 0.12 [22] 0.132 [23] 0.154 |
* 0.131 [12] 0.18 [13] 0.137 [14] 0.13 - - [20] 0.113 [21] 0.12 [22] 0.141 [23] 0.168 |
* 0.129 [12] 0.12 [13] 0.137 [14] 0.14 - - [20] 0.117 [21] 0.11 [22] 0.125 [23] 0.153 |
* 0.156 - [13] 0.14 - - - [20] 0.133 - - [23] 0.178 |
* 0.159 - - - - - [20] 0.137 - - [23] 0.222 |
* 0.149 - - - [15] 0.11 - [20] 0.12 - - [23] 0.16 |
| VTI |
* 0.08 [13] 0.30 - [16] 0.06 |
* 0.051 [13] 0.567 - - |
* 0.051 [13] 0.433 - - |
* 0.052 [13] 0.40 - - |
* 0.060 - - - |
* 0.059 - [15] 0.05 - |
| SPI |
* 4.75 [13] 7.72 - [16] 3.64 [22] 5.0 |
* 5.23 [13] 11.82 - - [22] 7.05 |
* 8.98 [13] 14.24 - - [22] 7.91 |
*11.39 [13] 22.42 - - - |
* 9.83 - - - - |
* 8.25 - [15] 9.80 - - |
| CPP (dB) |
* 9.701 - |
* 11.025 - |
* 9.818 - |
* 12.671 - |
* 14.075 - |
* 11.671 [17] 9.82 |
| CPP standard Deviation (dB) |
* 1.035 - |
* 0.938 - |
* 0.993 - |
* 1.501 - |
* 1.30 - |
* 1.166 [17] 0.96 |
| L/H ratio (dB) |
* 26.372 - |
* 30.361 - |
* 28.946 - |
* 31.810 - |
* 33.040 - |
* 30.390 [17] 36.47 |
| L/H ratio standard deviation (dB) |
* 2.245 - |
* 2.115 - |
* 2.159 - |
* 1.963 - |
* 2.034 - |
* 2.095 [17] 1.77 |
| CSID |
* 42.720 - |
* 29.869 - |
* 37.575 - |
* 24.851 - |
* 16.858 - |
* 29.264 [17] 21.84 |
Inspection of the data shows aspects of overall agreement for some parameters but there are notable differences for other parameters across studies, which are an obvious concern in validating a reference database. For the eventual purpose of specifying a normative value to be used clinically, a reasonable solution may be to take the upper bound of the values for each variable for those parameters in which pathology is indicated by a supra-normal value (as in the case with Jitt, Shim, and NHR). The variability across studies hinders an attempt to specify normative values that can be used with complete confidence. In addition, there are few consistent developmental trends across studies, with the exception of Fo, which shows decreasing values with age. The results of the present study are in agreement with [13] in showing age-related decreases in vFo and Jitt for both sexes.
Only limited comparison data for children have been published on several parameters included in the present study. Of particular interest are parameter values that show marked age differences for both sexes in this study. As expected, Fo decreased with age in both sexes beginning in most studies by the age of 7 to 9 years. An earlier study [11] of most of the same children reported on here, but that obtained Fo values from four different vowels produced in words yielded a similar result, namely that the mean values of Fo decreased for both sexes beginning at about 7 years of age, reaching adult values at about 14 for girls and 16 for boys. Other parameters showing fairly consistent changes with age in both sexes include vFo, Jitt, and PPQ, all of which are indices of variability in fundamental frequency, with higher values of these parameters in younger children probably reflecting immaturity in phonatory control. Spectral ratios such as SPI, L/H ratio, and CSID are influenced by the effects of age and sex on the distribution of energy across frequencies. Children have relatively more spectral energy in the higher spectral regions because of their higher fundamental and formant frequencies [11]. CSID, which is derived from the CPP, L/H ratio, and sex information through weighted multiple regression, showed a marked decrease (43% in females and 39% in males) from the youngest to oldest age groups. This decrease in CSID is consistent with Infusino et al. [25] who did not report specific CSID values for different ages but rather characterized overall changes in CSID across development.
In addition to the parameters reported in Tables 2, 3, and 4, MDVP provides data on parameters that do not apply to all voice samples. Degree of subharmonics, DSH (%), is an estimation of subharmonic to Fo components in the voice sample and is applicable only to voices that possess this feature. Subharmonics are small but distinct peaks located between two consecutive harmonic peaks in the power spectrum [26] and are thought to result from nonlinear properties in vocal fold vibration [27]. Table 6 shows the percentage of children for whom DSH was observed, along with the range of values greater than zero (i.e., data for children who exhibit subharmonics). DSH declines with age in the present study and therefore may reflect maturation of phonatory function, possibly because children learn to avoid this type of vocalization, or because of changes in vocal fold architecture, or both of these. However, DSH also could be a feature of a phonational register such as vocal fry, which may be adopted for sociocultural reasons [28,29].
Table 6.
Data for degree of subharmonics (DSH), showing for each age group the percentage occurrence and range of values other than zero. Data show decreasing DSH values as age increases.
| Group 1 4–6;11 yrs | Group 2 7–9;11 yrs | Group 3 10–12;11 yrs | Group 4 13–15;11 yrs | Group 5 16–19;11 yrs | |
|---|---|---|---|---|---|
| Females | 67 % 2.0 – 32.5 |
73 % 1.0 – 13.2 |
40 % 1.0 – 16.2 |
33 % 1.0 – 24.4 |
15 % 1.5 – 6.8 |
| Males | 67 % 3.4 – 21.2 |
47 % 1.0 – 4.3 |
33 % 1.2 – 14.3 |
20 % 3.2 – 11.3 |
0 NA |
The overall developmental pattern indicated by the data in the present study is that as children mature, they have decreasing values of Fo, vFo, Jitt, PPQ, L/H ratio standard deviation, and sex appropriate CSID. They also have increasing values of CPP standard deviation and L/H ratio, and are less likely to exhibit DSH.
4.0. Discussion.
Summaries of reference data in [4,5,6,7] and the present study provide information on MDVP and/or ADSV parameters across the lifespan in both sexes. These cross-sectional reference data from several different countries satisfy in part the recommendation by Brockmann-Bauser and Drinnan [25] that normative values should be revised to consider sex and vowel. The general conformity of the data for most parameters holds promise for the eventual establishment of a normative database suitable for clinical application, but there is considerable inter-individual variability at most ages and also variation in values reported in different studies. Parameters that had generally decreasing values with age in both females and males were: Fo, vFo, Jitt, PPQ, L/H standard deviation, and sex-appropriate CSID. Parameters that showed increasing values with age in both sexes were CPP standard deviation and L/H ratio. DSH decreased with age in both sexes. The parameters vFo and occurrence of DSH probably reflect maturation of vocal behavior per se, but sociocultural factors cannot be ruled out. The parameters L/H ratio and CSID likely are strongly influenced by the spectral differences resulting from growth of the vocal folds and the vocal tract, which point to the need for developmentally appropriate reference values. There is clearly a need for a developmental perspective and larger samples of children to establish a normative database, which would support clinical applications. As Infusino et al. [25] noted for the measure of CSID, “While all of the children included in the study had normal voices, younger children tended to display higher CSID values that would otherwise be labeled as dysphonic in adults” (p. 361). The data available at this time are not adequate to define a sex-specific normative pattern for all parameters of MDVP and ADSV. Most published reports focus on a small number of parameters, especially Jitt, Shim, and NHR. Some parameters, such as DSH, have been reported only infrequently but they may have value in describing vocal development [31] and disorders of phonation [32]. A profile of vocal function based on a combination of parameters may have utility for describing differences in both normal and pathological voice.
Consideration should also be given to other issues such as reliability of measures within a given analysis software and agreement between different software systems. Poor reliability has been reported for some MDVP measures [33,34], and further research is needed to determine the reliability of measures in both MDVP and ADSV for voices in both sexes across the lifespan. It is particularly concerning that reports indicate poor reliability for some of the most commonly used parameters (e.g., Jitt, Shim, NHR). However, other parameters may be more discriminating for clinical purposes. Al-Nasheri et al. [35] reported that the top three parameters for detection and classification of voice pathology in three databases were vAm, APQ, and Fo. In a meta-analysis of correlation coefficients between auditory-perceptual judgments and various acoustic measures, it was concluded that only a few acoustic parameters served as the best estimators for roughness and breathiness [36]. It remains to be determined if the same patterns apply to both children and adults.
The cross-sectional reference data in the present report pertain only to MDVP and ADSV applied to the analysis of a sustained vowel phonation. Generalization to other software systems should be done with caution if at all. Studies of agreement across software systems have revealed notable differences for some measures [6,37,38,39,40,41] and it cannot be assumed that a measure obtained with one software, such as MDVP or ADSV, is directly comparable to a similarly named measure in another software. At this point, it is preferable to establish normative databases and clinical guidelines specific to each analysis system. The MDVP data on children’s voice reported to date pertain almost entirely to children older than 4 years, but application to younger children, even newborns, is being reported [42].
5.0. Conclusion
Cross-sectional reference values are reported for children’s voice using the voice analysis systems MDVP and ADSV. The data for 158 children complement previously published data and therefore contribute to a cross-sectional reference database useful for research and for clinical practice related to children’s voice. A number of MDVP and ADSV measurements are reported for both sexes over the age range of 4 to 19 years. Data for MDVP and ADSV parameters are compiled from 12 previously published studies as well as the present study. These studies vary in the number of parameters reported, with the most frequently reported parameters being Fo, Jitt, Shim, and NHR.
What this study adds:
Reports data on a relatively large number of children to increase the reference pediatric database on acoustic measures of voice.
Reports data on an extensive set of variables from MDVP and ADSV, often used in research and clinical practice related to voice.
Compiles normative MDVP and ADSV data from 12 published sources.
6. Acknowledgments
This work was supported by supplement to NIH Research Grant R01 DC6282 (MRI and CT Studies of the Developing Vocal Tract, Houri K. Vorperian, Principal Investigator) from the National Institute on Deafness and other Communicative Disorders (NIDCD) and by a core grant U54 HD090256 to the Waisman Center from the National Institute of Child Health and Human Development (NICHHD). We thank Sophie Blankenheim for assistance with acoustic analysis.
Appendix 1.
MDVP parameters listed alphabetically as defined in instruction manual of Pentax Medical.
APQ (Amplitude Perturbation Quotient / %) - Relative evaluation of the period-to-period variability of the peak-to-peak amplitude within the analyzed voice sample at smoothing level of 11 periods. Voice break areas are excluded.
DSH (Degree of Sub-Harmonic Components / %) - Estimated relative evaluation of sub-harmonic to Fo components in the voice sample. The degree of sub-harmonic components in normal voices is expected to be equal to zero. DSH increases when double or triple pitch periods replace the fundamental period in certain segments over the analysis length.
Fo (Average Fundamental Frequency /Hz) - Average value of all extracted period-to-period fundamental frequency values. Voice break areas are excluded.
Jitt (Jitter Percent / %) – Relative evaluation of the period-to-period (very short-term) variability of the pitch within the analyzed voice sample. Voice break areas are excluded.
NHR (Noise-to-Harmonic Ratio) - Average ratio of the inharmonic spectral energy in the frequency range 1500–4500 Hz to the harmonic spectral energy in the frequency range 70–4500 Hz. NHR is a general evaluation of noise present in the analyzed signal.
PPQ (Pitch Period Perturbation Quotient / %) – Relative evaluation of the period-to-period variability of the pitch within the analyzed voice sample with a smoothing factor of 5 periods. Voice break areas are excluded.
Shim (Shimmer Percent / %). Relative evaluation of the period-to-period (very short-term) variability of the peak-to-peak amplitude within the analyzed voice sample. Voice break areas are excluded.
SPI (Soft Phonation Index) - Average ratio of the lower-frequency harmonic energy in the range 70–1600 Hz to the higher-frequency harmonic energy in the range 1600–4500 Hz. SPI can be considered to be indicator of how completely or tightly the vocal folds adduct during phonation.
STD (Standard Deviation of the Fundamental Frequency /Hz) - Standard deviation of all extracted period-to-period fundamental frequency values. Voice break areas are excluded.
vAm (Coefficient of Amplitude Variation /%) - Relative standard deviation of the peak-to-peak amplitude. It reflects in general the peak-to-peak amplitude variations (short to long-term) within the analyzed voice sample. Voice break areas are excluded.
vFo (Coefficient of Fundamental Frequency Variation /%) - Relative standard deviation of the fundamental frequency. It reflects, in general, the variation of Fo (short to long-term) within the analyzed voice sample. Voice break areas are excluded.
VTI (Voice Turbulence Index) - Average ratio of the spectral inharmonic high-frequency energy in the range 2800–5800 Hz to the spectral harmonic energy in the range 70–4500 Hz in areas of the signal where the influence of the frequency and amplitude variations, voice breaks and sub-harmonic components are minimal. VTI measures the relative energy level of high-frequency noise.
ADSV parameters listed alphabetically
CPP (cepstral peak prominence / dB) - Measures the difference between F0 intensity and the average intensity of the signal. A higher CPP should be seen with normal voices. Standard deviation of CPP is also calculated.
CSID (cepstral-spectral index of dysphonia) - A multivariate calculation that gives an estimation of vocal abnormality. Higher scores correlate with dysphonia. Sex appropriate values are calculated.
Fo – vocal fundamental frequency.
L/H ratio (low-to-high spectral ratio / dB) - The mean ratio of signal energy below 4000 Hz to the energy above 4000 Hz for the selected voice data frames. Standard deviation of this ratio is also calculated.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The authors have no conflicts of interest.
7 References
- [1].Kreiman J, Gerratt BR, Precoda K, Berke GS, Individual differences in voice quality perception. J Speech Lang. Hear. Res 35 (1992) 512–520. [DOI] [PubMed] [Google Scholar]
- [2].Oates J, Auditory-perceptual evaluation of disordered voice quality, Folia Phoniat. Logopaed 61 (2009) 49–56. 10.1159/000200768 [DOI] [PubMed] [Google Scholar]
- [3].Deliyski D, Effects of aging on selected acoustic voice parameters: Preliminary normative data and educational implications, Educ. Gerontol 27 (2001) 159–168. [Google Scholar]
- [4].Hema N, Mahesh S, Pushpavathi M, Normative data for Multi-Dimensional Voice Program (MDVP) for adults-A computerized voice analysis system, J. All India Inst. Speech Hear 28 (2009) 1–7. [Google Scholar]
- [5].Schaeffer N, Knudsen M, Small A, Multidimensional voice data on participants with perceptually normal voices from ages 60 to 80: a preliminary acoustic reference for the elderly population, J. Voice (2015) 631–637. 10.1016/j.jvoice.2014.10.003 [DOI] [PubMed] [Google Scholar]
- [6].Spazzapan EA, Marino VCDC, Cardoso VM, Berti LC, Fabron EMG, Acoustic characteristics of voice in different cycles of life: an integrative literature review. Revista CEFAC (2019) 21 10.1590/1982-0216/201921315018 [DOI] [Google Scholar]
- [7].Saggio G, Costantini G, Worldwide healthy adult voice baseline parameters: a comprehensive review. J. Voice (in press). [DOI] [PubMed] [Google Scholar]
- [8].Kempster GB, Gerratt BR, Abbott KV, Barkmeier-Kraemer J, Hillman RE, Consensus auditory-perceptual evaluation of voice: development of a standardized clinical protocol, Am. J. Speech-Lang. Pathol 18 (2009) 124–132. 10.1044/1058-0360(2008/08-0017) [DOI] [PubMed] [Google Scholar]
- [9].Awan SN, Roy N, Cohen SM, Exploring the relationship between spectral and cepstral measures of voice and the Voice Handicap Index (VHI), J. Voice, 28 (2014) 430–439. 10.1016/j.jvoice.2013.12.008 [DOI] [PubMed] [Google Scholar]
- [10].Peterson EA, Roy N, Awan SN, Merrill RM, Banks R, Tanner K, Toward validation of the cepstral spectral index of dysphonia (CSID) as an objective treatment outcomes measure, J. Voice 27 (2013) 401–410. 10.1016/j.jvoice.2013.04.002 [DOI] [PubMed] [Google Scholar]
- [11].Vorperian HK, Kent RD, Lee Y, Bolt DM, Corner vowels in males and females ages 4 to 20 years: Fundamental and F1-F4 formant frequencies. J. Acoust. Soc. Am 146 (2019) 3255–3274. 10.1121/1.5131271 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Abo-Ras YA, El-Maghraby R, Abdou RM, The normative study of acoustic parameters in normal Egyptian children aged 4–12 years. Alexandria J. Medicine, 49 (2013) 211–214. 10.1016/j.ajme.2013.02.001 [DOI] [Google Scholar]
- [13].Akmese PP, Kayikci MEK, Atas A, A. (2012). Acoustic characteristics of Turkish speaking children ages between 4 and 14 years old. J Int. Advances Otology, 8 (2012), 399. [Google Scholar]
- [14].Banik A, Arya S, Kant A, Vocal Parameters in Children between 4 To 12 Years of Age: An Attempt to Establish a Prototype Database. Intern. J. Scient. Research Publications, 11 (2015), 446–453. [Google Scholar]
- [15].Campisi P, Tewfik TL, Manoukian JJ, Schloss MD, Pelland-Blais E, Sadeghi N, Computer-assisted voice analysis: establishing a pediatric database. Arch. Otolaryngol–Head Neck Surg, 128 (2002) 156–160. https://doi:10.1001/archotol.128.2.156 [DOI] [PubMed] [Google Scholar]
- [16].Cappellari VM, Cielo CA, Vocal acoustic characteristics in pre-school aged children. Brazil. J. Otorhinolaryngol, 74 (2008), 265–272. 10.1016/S1808-8694(15)31099-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Diercks GR, Ojha S, Infusino S, Maurer R, Hartnick CJ, Consistency of voice frequency and perturbation measures in children using cepstral analyses: a movement toward increased recording stability. JAMA Otolaryngol.–Head Neck Surg, 139 (2013) 811–816. 10.1001/jamaoto.2013.3926 [DOI] [PubMed] [Google Scholar]
- [18].Hill CA, Ojha S, Maturo S, Maurer R, Bunting G, Hartnick CJ, Consistency of voice frequency and perturbation measures in children. Otolaryngol.--Head Neck Surg 148 (2013) 637–641. [DOI] [PubMed] [Google Scholar]
- [19].Jotz GP, Cervantes O, Settani FAP, Angelis ED, Acoustic measures for the detection of hoarseness in children. Int Arch Otorhinolaryngol, 10(1) (2006) 14–20. [Google Scholar]
- [20].Maturo S, Hill C, Bunting G, Ballif C, Maurer R, Hartnick C, Establishment of a normative pediatric acoustic database. Arch. Otolaryngol.–Head Neck Surg 138 (2012) 956–961. [DOI] [PubMed] [Google Scholar]
- [21].Okuda A, Tamai F, Shiromoto O, Acoustic parameters during phonation in children. Japan J Loped Phoniat 56 (2015) 166–170. [Google Scholar]
- [22].Tavares ELM, de Labio RB, Martins RHG, Normative study of vocal acoustic parameters from children from 4 to 12 years of age without vocal symptoms. A pilot study. Brazil. J. Otorhinolaryngol 76 (2010) 485–490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Toki EI, Tatsis G, Tafiadis D, Plachouras K, Ziavra N, Computer-based analysis of vocal acoustic parameters for typically developed Greek children. Ann. Otolaryngol. Rhinol 5 (2018) 1211. [Google Scholar]
- [24].Fitch WT, Giedd J, Morphology and development of the human vocal tract: A study using magnetic resonance imaging. J. Acoust. Soc. Am 106(3) (1999) 1511–1522. [DOI] [PubMed] [Google Scholar]
- [25].Infusino SA, Diercks GR, Rogers DJ, Garcia J, Ojha S, Maurer R,… Hartnick CJ, Establishment of a normative cepstral pediatric acoustic database. JAMA Otolaryngol.–Head Neck Surg 141 (2015) 358–363. 10.1001/jamaoto.2014.3545 [DOI] [PubMed] [Google Scholar]
- [26].Omori K, Kojima H, Kakani R, Slavit DH, Blaugrund SM, Acoustic characteristics of rough voice: subharmonics. J. Voice 10.1016/s0892-1997(97)80022-7 [DOI] [PubMed] [Google Scholar]
- [27].Herzel H, Steinecke I, Mende W, Wermke K, Chaos and bifurcations during voiced speech, in Mosekilde I, Mosekilde L (Eds.), Complexity, chaos, and biological evolution, Plenum Press, New York, 1991, pp. 41–50. [Google Scholar]
- [28].Cantor-Cutiva LC, Bottalico P, Hunter E, Factors associated with vocal fry among college students. Logoped. Phoniatr. Vocolog, 43(2) (2018), 73–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Dallaston K, Docherty G, The quantitative prevalence of creaky voice (vocal fry) in varieties of English: A systematic review of the literature. PloS one, 15(3), (2020), e0229960 10.1371/journal.pone.0229960 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Brockmann-Bauser M, Drinnan MJ, Routine acoustic voice analysis: time to think again? Current Opinion Otolaryngol. Head Neck Surg, 19 (2011) 165–170. 10.1097/MOO.0b013e32834575fe [DOI] [PubMed] [Google Scholar]
- [31].Fuamenya NA, N. A., Robb MP, Wermke K, K. (2015). Noisy but effective: crying across the first 3 months of life. J. Voice, 29(3) (2015), 281–286. [DOI] [PubMed] [Google Scholar]
- [32].Sakakibara KI, Imagawa H, Yokonishi H, Kimura M, Tayama N, N. Physiological observations and synthesis of subharmonic voices. In Proceedings of the Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (2011), pp. 1079–1085. http://www.apsipa.org/proceedings_2011/html/ [Google Scholar]
- [33].Carding PN, Steen IN, Webb A, Mackenzie K, Deary IJ, Wilson JA, The reliability and sensitivity to change of acoustic measures of voice quality. Clin. Otolaryngol 29 (2004) 538–544. 10.1111/j.1365-2273.2004.00846.x [DOI] [PubMed] [Google Scholar]
- [34].Gonzalez J, Cervera T, L Miralles J, Acoustic voice analysis: reliability of a set of multi-dimensional parameters. Acta Otorrinolaringol. Española, 53 (2002) 256–268. 10.1016/s0001-6519(02)78309-x [DOI] [PubMed] [Google Scholar]
- [35].Al-Nasheri A, Muhammad G, Alsulaiman M, Ali Z, Mesallam TA, Farahat M, Bencherif MA, An investigation of multidimensional voice program parameters in three different databases for voice pathology detection and classification, J. Voice, 31 (2017) 113–e9. 10.1016/j.jvoice.2016.03.019 [DOI] [PubMed] [Google Scholar]
- [36].Latoszek B. B. v., Maryn Y, Gerrits E, Bodt M. De, M. (2018). A meta-analysis: Acoustic measurement of roughness and breathiness. J. Speech Lang. Hearing Res, 61(2) (2018) 298–323. 10.1044/2017_JSLHR-S-16-0188 [DOI] [PubMed] [Google Scholar]
- [37].Freitas SV, Pestana PM, Almeida V, Ferreira A, Integrating voice evaluation: correlation between acoustic and audio-perceptual measures. J. Voice, 29 (2015) 390–e1. 10.1016/j.jvoice.2014.08.007 [DOI] [PubMed] [Google Scholar]
- [38].Maryn Y, Corthals P, De Bodt M, Van Cauwenberge P, Deliyski D, Perturbation measures of voice: a comparative study between Multi-Dimensional Voice Program and Praat. Folia Phoniat. Logopaed 61 (2009) 217–226. 10.1159/000227999 [DOI] [PubMed] [Google Scholar]
- [39].Richardson K, Matheron D, Martel-Sauvageau V, Vincent I, A comparative normative study between Multidimensional Voice Program, Praat, and TF32. Perspect. ASHA Special Interest Groups, 4 (2019) 563–573. 10.1044/2019_PERS-SIG19-2018-0006 [DOI] [Google Scholar]
- [40].Rohrer J, Maturo S, Hill C, Bunting G, Ballif C, Hartnick C, Pediatric voice analysis: comparison of 2 computerized analysis systems. JAMA Otolaryngol.–Head Neck Surg, 140 (2014) 742–745. 10.1001/jamaoto.2014.1162 [DOI] [PubMed] [Google Scholar]
- [41].Vaz-Freitas S, Pestana PM, Almeida V, Ferreira A, Acoustic analysis of voice signal: Comparison of four applications software. Biomed. Signal Process. Control 40 (2018) 318–323. 10.1016/j.bspc.2017.09.031 [DOI] [Google Scholar]
- [42].Devadiga DN, Remyasree T, Varghese AL, Ananthakrishna T, (2019). Acoustic analysis of infant cry using Multidimensional Voice Program–A preliminary study. Indian J. Public Health Res. Developm, 10(9) (2019) 430–434 [Google Scholar]