Effect of Voice Onset Type on Vocal Attack Time

*Ben C. Watson, *R. J. Baken, and Rick M. Roark, *yValhalla, New York

Summary: Vocal attack time (VAT) is the time lag between the growth of sound pressure (SP) and electroglottographic
(EGG) signals at vocal initiation. The characteristics of voice initiation are associated with issues of vocal hygiene, ef-
ficiency, and quality. Vocal onsets have commonly been qualitatively characterized into three types: hard, simultaneous,
and breathy. This study examines the effect of voice onset type on VAT values in normal speakers. SP and EGG record-
ings were obtained for 55 female and 57 male subjects while producing multiple tokens of three tasks (sustained // and
always as unaspirated onsets, and hallways as an aspirated onset). Results revealed a significant effect of onset type
on VAT, with the mean VAT for the hallways (aspirated) task greater than the mean VAT for the sustained // and
always (unaspirated) tasks. There was no significant VAT difference between the sustained // and always tasks.
Findings confirm the sensitivity of the VAT measure to vocal onset type and suggest its potential application as an objec-
tive and quantitative clinical measure of the type of vocal onset.
Key Words: Voice onsetVocal attack timePhonationVocal fold physiology.

INTRODUCTION the VAT measure against high-speed videoendoscopy, from

Voice researchers, teachers, and clinicians have long identified
voice initiation as an important aspect of vocal performance and
have tied it to issues of vocal hygiene,1,2 efficiency,3,4 and
quality.57 Indeed, the speed with which the vocal folds
adduct to the midline is considered an important variable in
the etiology of some voice disorders and a meaningful
indicator of neural dysfunction.814 It is generally accepted
that the initiation of phonation includes two somewhat
distinct phases: (1) the prephonatory adjustment phase that is
associated with setting the appropriate tension, gross
adduction, and aerodynamic forces15; and (2) the attack phase
that is associated with the onset of vocal fold oscillation and
sound generation. In one of the first voice studies to use
ultrahigh-speed cinematography, Moore16 addressed the issue
of the movements of the larynx leading to and including the
initiation of tone. Moore described the observed differences
between what he termed the glottal stroke or glottal shock
and the breathed and simultaneous vocal attacks. Based
largely on an auditory-perceptual categorization, voice onset
continues to be popularly divided into abrupt or hard, normal
or simultaneous, and breathy, aspirate, or soft glottal attack.17,18
Baken and Orlikoff19,20 proposed that the time delay between
the rise of the sound pressure (SP) and electroglottographic
(EGG) signals is related to characteristics of the attack
gesture. Specifically, a lag in the rise of the SP signal relative
to the EGG signal may correspond to a hard glottal onset
while a lead in the rise time of the SP signal may correspond
to an aspirate onset. The inter-signal lag, termed vocal attack
time (VAT), can be automatically and objectively estimated
using a cross-correlation method and is, therefore, free of oper-
ator/investigator bias and imprecision. Orlikoff et al21 validated
Accepted for publication December 8, 2014.
From the *Department of Speech-Language Pathology, New York Medical College,
Valhalla, New York; and the yDepartment of Otolaryngology, New York Medical College,
Valhalla, New York.
Address correspondence and reprint requests to Ben C. Watson, School of Health Sci-
ences and Practice, New York Medical College, 30 Plaza West, Valhalla, NY 10595. E-
mail: Watson@nymc.edu
Journal of Voice, Vol. 30, No. 1, pp. 11-14
0892-1997/$36.00
2016 The Voice Foundation
http://dx.doi.org/10.1016/j.jvoice.2014.12.004
which a digital kymogram was generated, and showed that
breathy onsets have positive VAT values, whereas hard
onsets have negative VAT values. Roark et al22 and Ma et al23
reported normative VAT values for English and Cantonese
speakers, respectively, in tasks for which voice onset was not
constrained.
The present study builds on the work of Roark et al22 to
examine effects of aspirate voice onsets on VAT values in
normal speakers. Specifically, we compare VAT values for tasks
that can be produced with either hard or simultaneous onsets
with a task that requires an aspirate initiation.
SUBJECTS
This study was approved by the Committee for the Protection of
Human Subjects (Institutional Review Board) of New York
Medical College, and all subjects provided informed consent
before data collection began. The subject sample for the present
study is the same sample previously described in Roark et al22
and included 55 females (mean age: 28 years; range:
2250 years) and 57 males (mean age: 29 years; range: 21
50 years). A health history that included chronic speech, voice,
cardiac, respiratory, or neurologic disorder was cause for exclu-
sion from the study. A current health status that included upper
respiratory infection was cause for a delay in participation until
the infection resolved.
METHODS
Simultaneous acquisition of SP and EGG signals was achieved
by use of a headphone-mounted microphone (model 333012;
RadioShack, Ft. Worth, TX) positioned approximately 5 cm in
front of the participants lips at approximately 15 off midline
and an EGG (Glottal Enterprises EG2; Syracuse, NY), respec-
tively. The SP and EGG signals were routed through an ampli-
fier and digitizer (M-Audio Fast Track; Irwindale, CA) and
digitized (44.1 ksamples/second; 16-bit resolution) to a laptop
computer for storage and subsequent analysis. Signal acquisi-
tion used Audacity software (v.3, General Public License,
http://audacity.sourceforge.net). Subjects sustained a vowel at
comfortable pitch and loudness while gain was adjusted to
12
FIGURE 1. Smoothed amplitude functions of the bandpass-filtered
sound pressure (SP) and electroglottographic (EGG) signals, yielding
ASP (t) and AEGG (t) respectively.
achieve adequate amplification without clipping while signals
were monitored on a two-channel oscilloscope.
The stimuli always and ahh (to represent the sustained
vowel, //) were used to elicit unconstrained, albeit typically
unaspirated, voice onsets while the task hallways was used
to elicit a specifically aspirate onset. The stimuli were printed
on individual 3 3 5 inch index cards (five cards per task).
The experimenter shuffled the cards and placed them face-
down in front of the subject. The subject was instructed to
turn over each card, silently read the task item, and then to
say the word at a comfortable pitch, loudness, and rate or to sus-
tain the vowel at a comfortable pitch and loudness for approx-
imately 2 seconds. Subjects were instructed that this was not a
reaction time task and to pause for several seconds between
cards. EGG and SP signals were recorded continuously until
the subject completed all 15 cards. The cards were then re-
shuffled, and one or two more trials were recorded. VAT data
for the vowel task were originally reported by Roark et al.22
SP and EGG signals were stored as stereo wave (.wav) files
and analyzed using custom algorithms implemented in MAT-
LAB (MathWorks, Natick, MA). The signal analysis process
is described in Roark et al22 in sufficient detail to replicate
this method and is summarized briefly here.
The data analysis process consists of four steps, each of
which includes graphical display and optional acoustic play-
back. Analysis was allowed to progress to the next stage only
if graphic display and/or acoustic playback met criteria for
acceptable data quality (eg, production of the correct task):
Step 1, signal verification: signals could be zoomed and
panned and/or played acoustically through external speakers
to verify accuracy of task target and fidelity of signal quality;
FIGURE 2. Difference functions, ASP0 (t) and AEGG0 (t) and the cross-correlation function. The time lag at the peak of cross-correlation provided
the vocal attack time, VAT.
Journal of Voice, Vol. 30, No. 1, 2016
Step 2, signal segmentation: automated identification of a
600 milliseconds segment of the SP and EGG signals
centered at the approximate time of vocal offset;
Step 3, F0-based frequency filtering and signal modeling:
automated extraction of a representative value of funda-
mental frequency (F0) of the raw EGG signal and band-
pass filtering the 600 milliseconds segment at 40% of the
F0 value. The analytic signal models of the band-pass
filtered SP signal and EGG signal are as follows:


XBP t AX tsin 2pfx tt ; (1)
where X is SP or EGG, and AX(t) is the instantaneous amplitude
and fX(t) is the instantaneous frequency of the modeled band-
passfiltered signals (Figure 1);
Step 4, extraction of measures: the instantaneous ampli-
tudes [AX(t), where X is SP or EGG] were smoothed and
difference functions [AX 0t AX t  AX t  Dt,
where X is SP or EGG and Dt 10 milliseconds] were
derived and normalized to represent local variation in the
SP and EGG amplitude models. The normalized cross-
correlation of the smoothed difference functions is shown
in Figure 2. Here, the peak of the correlation occurs at
9.98 milliseconds, which represents the VAT for this token
of hallways.
VAT measures were grouped by sex and task (always, // ,
hallways) and analyzed using a mixed model with sex and
task entered as fixed effects. A compound symmetric covariate
structure was used for the within subjects repeated measure
(token).
RESULTS
Following Roark et al,24 the linear correlation coefficient, Pear-
son r, computed for the smoothed SP and EGG difference func-
tions was used as the figure of merit (FOM) for measures of
VAT. An a priori criterion FOM of greater than 0.75 applied
to the database resulted in the rejection of 32 (2.8%) of the
1133 tokens for //, 41 of the 1086 tokens (3.7%) for always,
and 95 of the tokens 971 (9.7%) for hallways. Increased vari-
ability in the production of aspirate onsets likely led to more
frequent violation of assumptions underlying our FOM crite-
rion. There was no discernable pattern to the rejection of these
Ben C. Watson, et al Effect of Voice Onset Type on VAT 13

TABLE 1.
Summary Statistics of VAT (Milliseconds)
VAT

Group/ 2575
Subgroup N Mean (SD) Median Percentiles
Females, all 1512 2.66 (5.36) 2.82 0.59 to 6.31
tasks
Males, all 1510 3.55 (4.50) 3.67 0.72 to 6.34
tasks
FIGURE 3. Histograms of VAT measures by task (bin width 1 ms).
All subjects
// 1101 1.98 (5.32) 2.04 0.99 to 5.03
Always 1045 1.85 (4.89) 1.85 1.06 to 4.71
Hallways 876 6.01 (3.21) 5.78 3.97 to 8.19 DISCUSSION
Females Normative VAT values are reported for aspirated and unaspi-
// 519 0.78 (5.34) 0.93 1.81 to 3.83 rated voice onsets. We found no significant difference between
Always 535 0.94 (4.95) 1.08 1.83 to 4.05 females and males for VAT values collapsed across the three
Hallways 458 6.79 (3.08) 6.82 4.83 to 17.09 tasks. There was a significant effect of task on VAT. Adjusted
Males mean VAT values for the vowel and always tasks, both typi-
// 582 3.04 (4.90) 2.95 11.76 to 6.25 cally characterized by an unaspirated onset, differed by
Always 510 2.81 (4.65) 2.91 0.18 to 5.48 0.06 milliseconds. Both sexes showed longer VAT values for
Hallways 418 5.17 (3.14) 4.88 3.33 to 7.18 the hallways taska task phonetically constrained to be
Abbreviation: SD, standard deviation. initiated with an aspirated onset. However, females showed a
larger VAT difference between the unaspirated and aspirated
onsets (approximately 6 milliseconds) than did the males
tokens. As a consequence, the higher rejection rate was unlikely (approximately 2.2 milliseconds).
to bias the data set. This report is based on analysis of the re- Mean VAT for the hallways task in this study was shorter
maining 3022 tokens. than VAT values for the breathy onset condition reported by
Table 1 and Figure 3 summarize descriptive statistics for Orlikoff et al21 for both males (5.17 vs 26.4 milliseconds) and
VAT values by task. Table 2 summarizes the mixed model. females (6.79 vs 17.8 milliseconds). These differences may
The effect of sex was not significant. The effect of task was reside in the fact that subjects in the Orlikoff et al study inten-
significant (F[2, 199] 378.28, P < 0.0001). The adjusted tionally attempted to initiate voicing with a breathy onset. In so
mean VAT for the hallways task (5.99 milliseconds; stan- doing, they likely exaggerated the aspirate onset gesture.
dard error [SE]: 0.29 milliseconds) was greater than the The present findings reveal the sensitivity of the VAT mea-
adjusted mean VAT for the // (1.89 milliseconds; SE: sure to phonetically constrained differences in the voice onset
0.28 milliseconds) and the always (1.83 milliseconds; type. Furthermore, the finding that VAT values for hallways
SE: 0.29 milliseconds) tasks. There was no significant reported here were longer than VAT values for unaspirated on-
VAT difference between the // and always tasks. The sets and shorter than VAT values for the breathy onsets re-
Task by Sex interaction effect was also significant (F[2, ported by Orlikoff et al21 raises the possibility that VAT is
199] 88.26, P < 0.0001). Figure 4 shows both the high sensitive to the physiology of aspiration. That is, the range of
variability of VAT values and the nature of this interaction glottal constriction related to generating a sufficiently high
effect. Females showed smaller mean VAT values than Reynolds number is likely to influence the value of VAT. Sensi-
males for the // and always tasks, but a greater mean tivity of VAT to degree of glottal constriction supports its poten-
VAT value for the hallways task. tial as an objective and quantitative clinical measure of the type

TABLE 2.
Mixed Model of VAT, in Milliseconds: Testing the Fixed Effects of Task With Tokens as Random Effects
Subgroup Adjusted Mean SE of Adjusted Mean Mean Difference SE of Mean Difference t Value
// 1.89 0.28 Reference
Always 1.83 0.29 0.05 0.16 0.36
Hallways 5.98 0.29 4.10 0.17 24.01*
Always 1.83 0.29 Reference
Hallways 5.98 0.29 4.15 0.16 24.47*
*P < 0.0001.
14
FIGURE 4. Average (+/ 1 standard deviation) VAT measures by
task and sex.
of vocal onset. Future work will explore the sensitivity of VAT
to a wider range of voice onsets demonstrated in clinical popu-
lations. Another question to explore is how the variability of
VAT might impact its utility as a clinical measure. Data reported
here may, in fact, capture the normal range of variability. VAT in
clinical populations might exceed the range reported here.
Acknowledgments
The authors express their appreciation to Dr. Qiuhu Shi for his
assistance with the statistical analysis.
REFERENCES
1. Froeschels E. Hygiene of the voice. Arch Otolaryngol. 1943;38:122130.
2. Rees M. Harshness and glottal attack. J Speech Hear Res. 1958;1:344349.
3. Isshiki N, von Leden H. Hoarseness: aerodynamic studies. Arch Otolar-
yngol. 1964;80:206213.
4. Koike Y, Hirano M, von Leden H. Vocal initiation: acoustic and aero-
dynamic investigations of normal subjects. Folia Phoniatr. 1967;19:
173182.
5. Peters HFM, Boves L, van Dielen JCH. Perceptual judgment of the abrupt-
ness of voice onset in vowels as a function of the amplitude envelope. J
Speech Hear Disord. 1986;51:299308.
6. de Krom G. Some spectral correlates of pathological breathy and rough
voice quality for different types of vowel fragments. J Speech Hear Res.
1995;38:794811.
Journal of Voice, Vol. 30, No. 1, 2016
7. Steinhauer K, Grayhack JP, Smiley-Oyen AL, Shaiman S, McNeil MR. The
relationship among voice onset, voice quality, and fundamental frequency:
a dynamical perspective. J Voice. 2004;18:432442.
8. Luchsinger R, Arnold G. Voice-Speech-Language. Belmont, CA: Wads-
worth; 1965.
9. Koike Y, von Leden H. Pathologic vocal initiation. Ann Otol Rhinol Lar-
yngol. 1969;78:138148.
10. Leeper HA Jr. Voice initiation characteristics of normal children and chil-
dren with vocal nodules: a preliminary investigation. J Commun Disord.
1976;3:235238.
11. Lamprecht A. Untersuchungen zur Stimmschallparametern zu Beginn der
Phonation. Folia Phoniatr. 1990;42:302311. [German].
12. Pesak J, Urbanek K. Incoordination of phonation start in individuals with
balbuties. Folia Phoniatr. 1993;45:6267.
13. Andrade DF, Heuer R, Hockstein NE, Castro E, Spiegel JR, Sataloff RT.
The frequency of hard glottal attacks in patients with muscle tension
dysphonia, unilateral benign masses and bilateral benign masses. J Voice.
2000;14:240246.
14. Goberman AM, Blomgren M. Fundamental frequency change during offset
and onset of voicing in individuals with Parkinson disease. J Voice. 2008;
22:178191.
15. Faaborg-Andersen K. Electromyography of laryngeal muscles. Technics
and results. In: Current problems in phoniatrics and logopedics, vol 3.
Basel, Switzerland: Karger; 1965.
16. Moore P. Motion picture studies of the vocal folds and vocal attack. J
Speech Disord. 1938;3:235238.
17. Werner-Kukuk E, von Leden H. Vocal initiation: high speed cinemato-
graphic studies on normal subjects. Folia Phoniatr. 1970;22:107116.
18. Orlikoff RF, Kahane JC. Laryngeal structure and function. In: Lass NJ, ed.
Principles of Experimental Phonetics. St. Louis, MO: Mosby; 1996:
112181.
19. Baken RJ, Orlikoff RF. Vocal Fold Adduction Time Estimated From Glotto-
graphic Signals. Presented at the 25th Mid-Winter Meeting of the Associ-
ation for Research in Otolaryngology; St. Petersburg, FL; February 1998.
20. Baken RJ, Orlikoff RF. Estimating vocal fold adduction time from EGG and
acoustic records. In: Schutte HK, Dejonckere P, Leezenberg H, Mondelaers
B, Peters HF, eds. Programme and Abstract Book: 24th IALP Congress,
Amsterdam, The Netherlands; 1998:15.
21. Orlikoff RF, Deliyski DD, Baken RJ, Watson BC. Validation of a glotto-
graphic measure of vocal attack. J Voice. 2009;23:164168.
22. Roark RM, Watson BC, Baken RJ, Brown DJ, Thomas JM. Measures of
vocal attack time for healthy young adults. J Voice. 2012;26:1217.
23. Ma EPM, Baken RJ, Roark RM, Li PM. Effect of tones on vocal attack time
in Cantonese speakers. J Voice. 2012;26:670.e1670.e6.
24. Roark RM, Watson BC, Baken RJ. A Figure of Merit for Vocal Attack
Time. J Voice. 2012;26:811.