VR Public Speaking Simulations Can Make Voices Stronger and

More Effortful

Ïo Valls-Ratés

1,3 a

, Oliver Niebuhr

1 b

and Pilar Prieto

2,3 c

University of Southern Denmark, Alsion, 2, Sønderborg, Denmark

Universitat Pompeu Fabra, Roc Boronat, 138, Barcelona, Spain

Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig de Lluís Companys, 23, Barcelona, Spain

Keywords: Virtual Reality, Public Speaking, Prosody, Voice.

Abstract: In the field of public speaking, studies have mainly centered on the effects of virtual reality (VR) environments

in reducing public speaking anxiety (PSA). However, prior research on the effect of VR simulations on high-

school students' performance in terms of the prosody of their speech and number of gestures while being

immersed in a VR scenario is limited to just one study. The present paper examines the effects of practicing

speeches with a VR-simulated audience on self-perceived PSA, and speaking performance qualified on the

basis of the prosodic characteristics of the presenter’s voice and the rate of gestures they use while presenting.

Forty-seven high school students participated in either a VR group that practiced a two-minute speech in front

of a virtual audience, or a Non-VR group that delivered the same oral presentation alone in a room. Crucially,

these were compared with a baseline initial oral task where students presented in front of a live audience.

Practicing with VR resulted in significant differences across the groups pointing to VR-trained voices

becoming stronger, more effortful and louder. Simulated audience seems to help speakers develop more

audience-oriented prosody. This is particularly useful for rehearsing public speaking skills in the context of

secondary school education to improve students' oral competence.

1 INTRODUCTION

Practicing public speaking in classrooms is crucial for

students' confidence and social development (King,

2002; Iberri-Shea, 2009). It enhances skills such as

decision-making, critical thinking, and empathy

(Schneider et al., 2017). This skillset is vital for future

professional success (Nguyen, 2015) and can

alleviate public speaking anxiety (Liao, 2014).

Educational institutions should recognize its

importance in fostering self-confidence and self-

directed learning (Munby, 2011), while instructors

play a key role in motivating student engagement

(Kaufmann & Tatum, 2017).

2 RELATED WORK

Against this background, the aim of this study was to

https://orcid.org/0000-0001-9511-6927

https://orcid.org/0000-0002-8623-1680

https://orcid.org/0000-0001-8175-1081

investigate the effects of VR on adolescents' voice

and gesture, and their self-perceived anxiety while

being immersed in a VR setting performing a speech.

Public-speaking training is rarely offered in

educational settings. This is unfortunate, given that,

according to Ford and Wolvin (1993), once public

speaking is trained in the classroom, students

perceive that their communication becomes more

effective and they feel more self-confident, more

confident that they are well-regarded by others, more

able to reason with other people and more skillful at

using language appropriately. The ability to

communicate effectively and appropriately is learned

and, therefore, must be taught (Morreale et al., 2000).

In short, teaching public speaking skills needs to be

directly integrated into the classroom.

Mastering public speaking shares notable parallels

with the process of acquiring a new language. Research

suggests a critical time window exists for language

acquisition, as demonstrated by studies (e.g.,

Valls-Ratés, Ï., Niebuhr, O. and Prieto, P.

VR Public Speaking Simulations Can Make Voices Stronger and More Effortful.

DOI: 10.5220/0012748900003693

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 16th International Conference on Computer Supported Education (CSEDU 2024) - Volume 1, pages 685-693

ISBN: 978-989-758-697-2; ISSN: 2184-5026

685

Hartshorne et al., 2018). This critical period, shown to

extend to aspects such as prosody—the rhythmic,

intonational, and stress patterns defining a language

(Huang & Jun, 2011)—underscores the importance of

timely exposure and practice in developing linguistic

fluency. Just as proficiency in a new language benefits

from early immersion, effective public speaking skills

thrive with deliberate training and experience,

emphasizing the need for strategic interventions to

optimize learning outcomes, especially in the

classrooms.

A technology that can help engaging students in

the learning process is virtual-reality (henceforth

VR). VR technology displays three-dimensional

computer-generated scenes which create the illusion

for the viewer that he or she is physically located

within that simulated space and interacting with it, in

other words, a sensation of physical presence

(Radianti et al., 2020). VR has been used in a wide

variety of fields to treat phobias and post traumatic

disorder (Baños et al., 2011), in the entertainment

industry (Bianchi-Berthouze, 2013) and for medical

rehabilitation purposes (Bourdin et al., 2019). Several

studies have also assessed VR as a tool to be included

in speech communication courses to enhance

students’ performance and make learning more

meaningful, and to reduce public-speaking anxiety

(henceforth PSA). For instance, in a meta-analysis

exploring PSA and VR, Hui Lim et al. (2022)

identified a total of 92 studies and they concluded that

since the results of using VR were similar to those

obtained by other modes of therapy, it can be

considered an effective tool to treat clinical PSA (see

also Daniels et al., 2020) and appropriate as a

complementary method to other therapies such as

cognitive behavior therapy. Indeed, other studies

have shown how combining VR with other modalities

of therapy results in successful outcomes (e.g.,

Anderson et al., 2005; Wallach et al., 2009).

Very few studies have investigated the effects of

VR on high school students' public speaking

performance and anxiety (Kahlon et al., 2019; Valls-

Ratés et al., 2022; Valls-Ratés et al., 2023); and to our

knowledge only one has explored adolescents' vocal

and gestural patterns/behavior while being immersed

in a VR scenario (see Valls-Ratés et al., 2021).

The aim of the present study was to evaluate

adolescents' vocal and gestural patterns in virtual

scenarios. Studies that have analyzed the effects of

VR on prosody have been conducted in higher-

education (e.g., Niebuhr & Michalsky, 2018; Notaro

et al., 2021) and results show that VR can be an

effective tool to develop more audience-oriented

voices while performing a speech, pointing to a

potential effect of VR-assisted public-speaking

practice on not only the self-assessed anxiety of the

participants but also their delivery style in terms of

prosody and gesture use while practicing their

speeches. To this end, in a mixed-model experimental

design, high school students were asked to practice

their speeches either in front of a VR audience

(experimental condition) or alone in a classroom

(control condition). The two conditions served as our

between-subjects independent variable. Importantly,

in order to have reference values in the three domains

of interest, namely anxiety, prosody and gesture use,

before rehearsing in one of the two conditions each

speaker performed a short oral speech in front of a

real audience of three people. The real vs. VR

audience was a further within-subjects independent

variable in our experiment.

We hypothesized that practicing speeches within

VR settings would be conducive to (1) self-reports of

higher levels of self-perceived anxiety in comparison

to practicing alone without VR and (2) a more

audience-oriented nonverbal communication style in

terms of prosodies and gestures. In order to address

the second part of the hypothesis, a comprehensive

analysis of the 21 prosodic characteristics of the

target speeches including pitch (i.e., f0), tempo and

voice quality. Additionally, speakers’ gesture rate

during their public-speaking performance was

determined.

3 METHOD

3.1 Participants, Experimental Design,

Materials and Procedure

A total of 47 secondary school students from Barcelona

participated in the study. The mean age of participants

(67.18% female; 32.82% male) was 16.45 years (SD =

0.36). The experiment consisted of a mixed-model

experimental design whose main factor was a between-

subjects variable with the two conditions VR and Non-

VR, see Figure 1. First, all the participants took part in

a one-hour initial information session. Second, they

performed two public speaking tasks, namely (a) a

baseline two-minute public speaking task in front of a

live audience and (b) a two-minute public speaking

task, performed under one of two conditions, either in

front of a VR-simulated audience (VR PRACTICE), or

speaking alone (NON-VR PRACTICE). Following the

information session, the experimenter randomly

divided participants into two groups, the VR group (n

= 27) and the Non-VR group (n = 20).

ERSeGEL 2024 - Workshop on Extended Reality and Serious Games for Education and Learning

686

Figure 1: Experimental design.

3.1.1 Life Audience Speech (Baseline)

One week after the information session, participants

were given written instructions on how to prepare

what they planned to say, starting from the topic

“Adolescents need to spend more time in nature” –

and a list of five arguments that they could use and

elaborate on in their speech. They were allotted two

minutes for preparation and were then proceeded to

the adjacent room where their audience was waiting.

They were allowed a maximum of two minutes to

deliver their speech.

3.1.2 VR and Non-VR Practice Sessions

The practice session took place one week after

participants spoke to the live audience. The procedure

for the practice was the same as for the speech to the

live audience except that in this case the topic was

“The house of my dreams" and they were offered a

set of five questions instead of arguments to help

them prepare the presentation. Again, after two

minutes of preparation, they were accompanied to an

adjacent classroom. At this point, however, the

procedure followed diverged, depending on the group

to which participants had been allocated, VR or Non-

VR. Participants in the VR practice group were fitted

with a Clip Sonic

VR headset, to which a

smartphone was attached. A VR interface application

installed on the smartphone called BeyondVR

simulated a stage and gave the headset wearer the

illusion that they were standing in front of an

audience, see Figure 2. This virtual audience made

small realistic movements while seated, and it

conveyed an attentive attitude by making eye contact

with the speaker and, more generally, signaling their

interest in what the speaker was saying. These

realistic features were intended to make the audience

seem believably real and enhance the headset

wearer’s sense of presence (Slater et al., 1999). VR

group participants were able to monitor their speaking

time by referring to a timer displayed in their field of

vision by the headset. For Non-VR group

participants, the procedure was the identical, except

that they gave their speech alone in the classroom

without any VR equipment. However, they had

access to their speaking time on a computer screen

placed close to them. The performance of all

participants was video-recorded.

Figure 2: Screenshot of the VR scenario generated by

BeyondVR that was seen by VR group participants.

3.1.3 Self-Assessed Anxiety

Just prior to performing the two public speaking

tasks, each participant completed the Subjective Units

of Distress Scale (SUDS) (Wolpe, 1969) form to

indicate their level of anxiety. This instrument (a

long-standing and validated questionnaire (see Thoen

et al., 2023) yields a score from 0 (total relief) to 100

(the highest fear ever experienced). The participant

was told "Please rate your level of distress from 0 to

100" and asked to read the descriptors for each 0-100

value in order to quantify their overall distress.

3.2 Data Analysis

Because each of the 47 participants delivered two

speeches, one to the live audience and one in the

practice session, a total of 94 recordings were

obtained for analysis.

3.2.1 Prosodic Measures

Acoustic-prosodic analysis of the audio tracks of all

94 speeches was performed automatically by means

of the ProsodyPro script by Xu (2013) and the

supplementary analysis script by De Jong and Wempe

(2009), both using the (gender-specific) default

PRAAT settings (Boersma & Weenink, 2007). The

analysis included a total of 21 different prosodic

parameters, namely five f0 parameters, seven

duration parameters and nine voice quality

parameters. For a detailed explanation of the

measured parameters see Liu & Xu, 2014.

3.2.2 Manual Gesture Rate

All manual communicative gestures present in the

speakers’ speeches (to the live audience speech and in

the practice speeches in the two conditions) were

VR Public Speaking Simulations Can Make Voices Stronger and More Effortful

687

annotated. Following the M3D approach (see Rohrer et

al., 2021 for more details on the procedure), we

considered each manual stroke (the most effortful part

of the gesture that usually constitutes its semantic unit;

Kendon, 2004; McNeill, 1992) as corresponding to a

manual gesture. Non-communicative gestures such as

self-adaptors (e.g., scratching, touching hair; Ekman &

Friesen, 1969) were disregarded. For every speech, the

overall gesture rate was calculated as the total number

of manual gestures produced relative to the phonation

time in minutes (gestures/phonation time).

3.3 Statistical Analyses

Statistical analyses were performed using IBM SPSS

Statistics version 19. A total of 23 of GLMMs were

run for each of the following dependent variables:

self-perceived anxiety (SUDS), a set of 21 values for

all the prosodic parameters (five for f0, seven for

duration and nine for voice quality), and the gesture

rate. All the GLMM models included Condition (two

levels: VR vs. Non-VR, between-subjects factor) and

Time (two levels: Live Audience Speech; Practice

Session) and their interactions as fixed factors.

Subject was included as a random factor. Pairwise

comparisons and post-hoc tests were carried out for

the significant main effects and interactions.

4 RESULTS

4.1 Self-Assessed Anxiety

Figure 3: Mean SUDS values prior to the live audience

speech and practice session for both VR and Non-VR

conditions.

The GLMM analysis for SUDS showed a main effect

of Condition (F(1,88) = 13.513, p < .001) that

indicated that the participants of the VR group

displayed significantly higher values than the Non-

VR group, and not only for the practice speech but

also when both groups were compared, respectively,

to their live audience speeches (β = 13.942, SE =

3.793, p < .001). The analysis also showed a main

effect of Time (F(1,88) = 38.796, p < .001), meaning

that the SUDS anxiety values obtained prior to the

live audience speech (LAS) were significantly higher

than those obtained prior to the practice session (PS)

(β = 20.712, SE = 3.325, p < .001). Figure 3 shows

mean SUDS scores separated by Condition (VR;

Non-VR) and Time (Live Audience Speech = LAS;

Practice Session = PS).

4.2 Prosodic Parameters

4.2.1 F0 Domain

Regarding the f0 domain, five GLMMs were applied

to our target variables, namely minimum and

maximum f0, f0 variability (in terms of the standard

deviation), mean f0 and f0 range. Table 1 shows a

summary of those GLMM analyses in terms of main

effects (Time and Condition), as well as interactions

between Time and Condition of the most relevant

variables. Summarizing, a main effect of Time was

obtained for f0 min, f0 variability and mean f0,

showing that at Live Audience Speech values were

higher for f0 min and mean f0, but significant lower for

f0 variability. A main effect of Condition was obtained

for f0 min, f0 max and mean f0, showing higher f0

values in the VR condition for the three variables.

However, no significant interactions were obtained.

Table 1: Summary of the GLMM analyses for the f0

variables in terms of main effects and interactions (ø means

no significant effect, whereas all non-ø table cells refer to

significant differences between the paired conditions at

p<0.05. Live Audience Speech = LAS; Practice Session =

PS).

Variable

Main effect of

Time

Main effect of

Condition

Interaction

Time*Condition

f0 min

LAS



Non-VR

f0 max



Non-VR

mean f0

LAS



Non-VR

4.2.2 Tempo Domain

With regard to the tempo domain, seven GLMMs

were applied to each of the target dependent

variables, namely total number of syllables, total

number of silent pauses, total time of the presentation,

total speaking time, the speech rate, the net syllable

rate and ASD. Table 2 shows a summary of these

GLMM analyses in terms of main effects (Time and

ERSeGEL 2024 - Workshop on Extended Reality and Serious Games for Education and Learning

688

Condition), as well as interactions between Time and

Condition of the most relevant variables.

Summarizing, no main effects of Time were obtained.

A main effect of Condition was obtained for three

variables: speech rate, net syllable rate and ASD,

meaning that the participants in the VR group had

significantly higher values for speech rate and net

syllable rate values, and lower ASD values than the

Non-VR group. However, no significant interactions

were obtained.

Table 2: Summary of the GLMM analyses for the seven

duration variables, in terms of main effects and interactions.

4.2.3 Voice Quality Domain

Table 3: Summary of the GLMM analyses for the nine

voice variables in terms of main effects and interactions.

Variable

Main effect

of Time

Main effect

of Condition

Interaction

Time*Condition

h1*-h2*



Non-VR



Non-VR

h1-A3



Non-VR



Non-VR

CoG

LAS





Non-VR

Formant

dispersion 1-3

Non-VR



Shimmer



Non-VR



Non-VR

Jitter



Non-VR



Non-VR

Hammarberg

LAS





Non-VR



Non-VR

In the domain of voice quality measurements, nine

GLMMs were applied to the nine target variables,

namely h1*-h2*, h1-A3, CPP, Harmonicity, CoG,

formant dispersion 1-3, shimmer, jitter, and

Hammarberg index. Table 3 shows a summary of

those GLMM analyses in terms of main effects (Time

and Condition), as well as interactions between Time

and Condition of the most relevant variables.

Summarizing, a main effect of Time was obtained

for two variables, namely CoG and Hammarberg

index, meaning that at baseline (Live Audience

Speech) values were lower for CoG and higher for

Hammarberg index. A main effect of Condition was

Figure 4: Mean voice quality values (namely h1*-h2*, h1-

A3, Hammarberg index, jitter and shimmer) in the Live

Audience Speech and the Practice Session for the variables

that obtained a significant interaction Time * Condition, for

both VR and Non-VR conditions.

obtained for seven variables, namely h1*-h2*, h1-A3,

CoG, formant dispersion 1-3, shimmer, jitter and

Hammarberg index, meaning that the participants in

the VR group obtained higher values compared to the

Non-VR group, in both the Live Audience Speech

and the Practice Session, except for formant

Variable

Main effect of

Time

Main effect

of Condition

Interaction

Time*Condition

Number of

silent pauses

Speech rate

LAS



Non-VR

VR Public Speaking Simulations Can Make Voices Stronger and More Effortful

689

dispersion 1-3. Significant interactions were obtained

for h1*-h2*, h1-A3, shimmer, jitter and Hammarberg

index, showing higher values for the VR condition for

all the variables. The graphs in Figure 4 show the

mean voice quality values that obtained a significant

interaction Time * Condition, for both VR and Non-

VR conditions.

4.2.4 Manual Gesture Rate

A GLMM was applied for manual gesture rate. A

main effect of Time was obtained (F(1,84) = 40.601,

p < .001), showing that Live Audience Speech scores

were higher across groups (β = 16.410, SE = 2.575, p

< .001). However, no interaction was obtained

between Time and Condition. The graph in Figure 5

shows the mean gesture rate values for both VR and

Non-VR conditions.

Figure 5: Mean gesture rate values for the Live Audience

Speech and Practice Session for both VR and Non-VR

conditions.

5 DISCUSSION AND

CONCLUSIONS

The present study was designed to determine the

effects of practicing a short oral presentation to an

artificial VR-generated audience as compared to

practicing alone in a classroom on self-perceived

anxiety and a comprehensive set of prosodic features,

together with gesture rate measures. Forty-seven

high-school students participated in this mixed-model

experiment. In order to obtain a baseline measure, all

participants were asked to perform a speech in front

of a live audience before performing the actual

experimental task.

With respect to the first research question, it was

found that the self-reported anxiety levels decreased

significantly for both groups from the live audience

speech to the practice session. No significant

interactions were found, meaning that the two

conditions were not different with respect to the

baseline. The lack of effect on self-reported anxiety

measures could be explained by two reasons. First,

SUDS self-reports were the only measure of

participants' distress arousal prior to the performance

of public speaking. Adding other questionnaires and

combining them with physiological measures would

have allowed us to obtain a more fine-grained picture

of the distress and anxiety levels as the speakers faced

and talked to live and virtual audiences. Also, it is

conceivable that higher stress levels would be

reported after participants had put on the headset and

were facing the virtual audience than those reported

before the headset was put on.

Turning to the prosodic analysis of the

participants’ speeches, and to the tempo domain at

first, no significant interactions were obtained for any

of the parameters (total number of syllables, total

number of silent pauses, total time of the presentation,

total speaking time, speech rate, net syllable rate and

ASD), meaning that both VR and Non-VR groups

showed a similar (change in) behavior when their

practice speeches are compared to their live audience

speeches. In the f0 domain, a main effect of time was

obtained for f0 min, f0 variability, mean f0 and range,

and a main effect of condition for f0 min, f0 max, and

mean f0. The main effect of Condition revealed that

f0 min, f0 max and mean f0 were higher for the VR

condition than for the Non-VR condition. These

results are in line with those reported by Niebuhr and

Michalsky (2018), Notaro et al. (2021) and Remacle

et al. (2020), who found higher f0 levels and higher-

level melodic variation when participants were

immersed in artificial VR environments. In Niebuhr

and Michalsky, however, there is no comparison to a

speech performed in front of a live audience, unlike

in the other two studies (Notaro et al., 2021 and

Remacle et al., 2020) as well as the present

study. However, we see that all f0 parameters

maintain the same high levels in the live audience

speech as in the practice session. Thus, no significant

interaction was found here between time and

condition, meaning that there was no significant

difference between the f0 characteristics of speech in

the VR and the Non-VR conditions relative to the f0

features of the baseline speech condition in front of

an audience.

Importantly, the main difference between this

study and previous ones is the fact that our study did

not include any form of feedback during or after the

VR practice. Providing feedback on nonverbal

aspects of a speaker’s performance seems to be

fundamental to achieving improvement, as VR per se

does not include this feature in an automatic way

(e.g., Niebuhr & Michalsky, 2018). Importantly, our

ERSeGEL 2024 - Workshop on Extended Reality and Serious Games for Education and Learning

690

study did not find a significant difference between the

VR and non-VR conditions in the self-assessment

measure of anxiety, as well as in the duration and f0

measures of practice speeches. However, even though

this result distinguishes our study from some previous

findings, it needs to be related to the fact that we have

taken into account baseline measures. We believe that

a between-subjects factor involving VR-assisted and

non-VR-assisted conditions must be checked against

such within-subjects baseline condition where

participants perform a speech before a live audience

in order to assess potential individual differences.

Crucially, in the voice quality domain (h1*-h2*,

h1-A3, CPP, Harmonicity, CoG, formant dispersion

1-3, shimmer, jitter, and Hammarberg index),

significant interactions were obtained between time

and condition. Specifically, VR-assisted speakers

tended to use a louder and more powerful voice than

Non-VR-assisted speakers relative to when they were

addressing a live audience speech. These results are

consistent with Niebuhr & Michalsky (2018) and

Remacle et al. (2020). Practicing with VR also was

reflected in a higher Hammarberg index and h1-A3,

which suggests a more effortful and aroused voice

quality (Niebuhr & Taghva, 2022; Tamarit et al.,

2008). VR speeches showed a significant decrease for

shimmer, that is, less shaky, nervous, stressful voice,

whereas the opposite was found for the Non-VR

speeches. By contrast, Non-VR participants

significantly increased in jitter during their practice

session speeches, showing a less harmonic, tenser or

creaky voice.

Regarding the effects of manual gesture rate,

when comparing the amount of gesture produced in

front of the live audience compared to the practice

session, there was a significant reduction in the

gesture rate for both groups. That is, VR and Non-VR

participants both produced gestures less often in the

practice sessions. This decrease was greater in the

Non-VR condition, albeit not significantly so. These

results are in line with Notaro et al. (2021). Following

their reasoning, a possible explanation for this might

be that participants who are immersed in VR and

wearing a headset cannot see their own hands. For

Non-VR group participants, the reason was probably

different; here, the decrease may be due to an absence

of motivation and engagement because they are alone

in a classroom, giving a speech to no one. Taking

these results into account, we cannot confirm the

hypothesis that predicted an increase in gesture rate

in the VR condition, as the tendency was very similar

across the two groups.

In summary, participants that practiced their short

speeches within an unsupervised VR environment in

front of a virtual audience had in effect a more

realistic experience (in line with Selck et al., 2022).

As a result, regarding prosodic parameters, we see an

increase in vocal effort and loudness, with voices that

are stronger (hence also less shaky and stressed) and

aroused, which reflects a more audience-oriented

manner of speaking. The presence of the virtual

audience made participants more engaged and

encouraged them to use their voices similarly to how

they would have done in front of a live audience.

Some limitations of the study must be noted. First,

the study is based on a relatively small sample size

(for having a between-subjects factor involved). So,

results cannot be generalized to other age groups or

clinical populations. Second, assessing gesture rate

might not be enough to differentiate between the

gestural behavior of participants in both conditions.

Adding a more complete assessment of overall

multimodal behavior including body movement,

facial expressions, eye contact, types of hand gesture,

and so on could expand our knowledge about the

participants’ body engagement during VR

experiences. Finally, the addition of feedback about

speaker performance during or after public speaking

performances could have also favored the public

speaking VR experience of these young students, as

feedback has been shown to be valuable for learning

and skills improvement (Van Ginkel et al., 2019;

King et al., 2000).

There is abundant room for further progress in

determining the effects of VR immersion in students'

overall performances not just in one practice session,

but in several, to explore the differences between the

first VR immersion and the subsequent ones. Also,

not only comparting VR environments to being alone

in a classroom, but also comparing it to other ways of

rehearsing a speech. Longer VR sessions as well as

longer time to prepare the speeches could be also

further explored.

All in all, the present study highlights the value of

using VR for public speaking practice in secondary

school settings. If the current trend is for educational

policies to promote the learning of public speaking

skills, then opportunities should be provided for

students to rehearse their presentations and speeches

using virtual environments. In our view, combining

VR immersion with other sorts of training in the

classroom to develop related skills such as quality

conversation, active listening and critical thinking can

be key to broadening students' competence in both

their daily and future professional lives, in a more

engaging and fun way.

Future research should focus on analyzing the

long-term effects of virtual simulations on real

VR Public Speaking Simulations Can Make Voices Stronger and More Effortful

691

environments. Longitudinal studies could assess

students' perceptions of enjoyment and usefulness

with VR to determine if their high regard for it

persists beyond novelty. Additionally, extending

training sessions, duration, and incorporating

feedback strategies are essential future aims in both

research and practice.

Communicating effectively is an ability that needs

to be developed and mastered through the years. The

earlier we start teaching these skills, the easier it will

be to apply them properly in any communicative

situation. By nurturing these skills early on, educators

and parents empower young learners to communicate

effectively, collaborate productively, and navigate

the world with confidence and competence.

REFERENCES

Anderson, P. L., Zimand, E., Hodges, L. F., & Rothbaum,

B. O. (2005). Cognitive behavioral therapy for public-

speaking anxiety using virtual reality for exposure.

Depression and Anxiety, 22(3), 156-158.

Baños, R. M., Guillen, V., Quero, S., García-Palacios, A.,

Alcañiz, M., & Botella, C. (2011). A Virtual Reality

System for the Treatment of Stress-Related Disorders:

A Preliminary Analysis of Efficacy Compared to a

Standard Cognitive Behavioral Program. International

Journal of Human-Computer Studies, 69(9), 602-613.

Bianchi-Berthouze, N. (2013). Understanding the Role of

Body Movement in Player Engagement, Human-

Computer Interaction, 28(1), 40-75.

Boersma, P., & Weenink, D. (2007). PRAAT: Doing

Phonetics by Computer (Version 5.2.34). [Computer

software].

Bourdin, P., Martini, M., & Sanchez-Vives, M. V. (2019).

Altered visual feedback from an embodied avatar

unconsciously influences movement amplitude and

muscle activity. Scientific Reports, 9(1), [19747]. doi:

10.1038/s41598-019-56034-5

Daniels, M. M., Palaoag, T., & Daniels, M. (2020). Efficacy

of virtual reality in reducing fear of public speaking: A

systematic review. In International Conference on

Information Technology and Digital Applications,

803(1). doi: 10.1088/1757-899X/803/1/012003

De Jong, N. H., & Wempe, T. (2009). Praat script to detect

syllable nuclei and measure speech rate automatically.

Behavior research methods, 41(2), 385-390.

Ekman, P., & Friesen, W.V. (1969). The Repertoire or

Nonverbal Behavior: Categories, Origins, Usage and

Coding. Semiotica, 1, 49-98. doi:

10.1515/semi.1969.1.1.49

Ford, W. S. Z., & Wolvin, A. D. (1993). The differential

impact of a basic communication course on perceived

communication competencies in class, work, and social

contexts. Communication Education, 42, 215-233.

Hartshorne, J. K., Tenenbaum, J. B., & Pinker, S. (2018). A

critical period for second language acquisition:

Evidence from 2/3 million English speakers, Cognition,

177, 263-277. doi:10.1016/j.cognition.2018.04.007.

Huang, B. H., & Jun, S. A. (2011). The Effect of Age on the

Acquisition of Second Language Prosody. Language

and Speech, 54(3), 387-414. doi:

10.1177/0023830911402599

Hui Lim, M., Aryadoust, V., & Esposito, G. (2022). A

meta-analysis of the effect of virtual reality on reducing

public speaking anxiety. Current Psychology. Advance

online publication. doi: 10.1007/s12144-021-02684-6

Iberri-Shea, G. (2009). Using Public Speaking Tasks in

English Language Teaching. English Teaching Forum,

47(2), 18-23.

Kaufmann, R., & Tatum, N.T. (2017). Do we know what

we think we know? On the importance of replication in

instructional communication research. Communication

Education, 66, 479-481.

Kahlon, S., Lindner, P., & Nordgreen, T. (2019). Virtual

reality exposure therapy for adolescents with fear of

public speaking: A non-randomized feasibility and pilot

study. Child and Adolescent Psychiatry and Mental

Health, 13(47), 1-10.

Kendon, A. (2004) Gesture: Visible Action as Utterance.

Cambridge University Press, Cambridge.

King, P., Young, M., & Behnke, R. (2000). Public speaking

performance improvement as a function of information

processing in immediate and delayed feedback

interventions. Communication Education, 49(4), 365-

374. doi: 10.1080/03634520009379224

Liao, H. (2014). Examining the Role of Collaborative

Learning in a Public Speaking Course. College

Teaching, 62(2), 47-54, doi:

10.1080/87567555.2013.855891

Liu, X. & Xu, Y. (2014). Body size projection by voice

quality in emotional speech-Evidence from Mandarin

Chinese. Proceedings of the International Conference

on Speech Prosody, 974-977.

McNeill, D. (1992). Hand and mind: What gestures reveal

about thought. University of Chicago Press.

Morreale, S. P., Osborn, M. M., & Pearson, J. C. (2000).

Why communication is important: A rationale for the

centrality of the study of communication. Journal of the

Association for Communication Administration, 29(1),

1-25.

Munby, I. (2011). The oral presentation: An EFL teachers'

toolkit. Studies in Culture, 99, 143-168.

Nguyen, H. (2015). Student perceptions of the use of Pecha

Kucha presentations for EFL reading classes. Language

in Education in Asia, 6(2), 135-149.

Niebuhr, O. & Taghva, N. (2022, in review). Posture and

Gestures can affect the prosodic speaker impact in a

remote presentation. In M. Oliveira (Ed.), Prosodic

Interfaces. Berlin/New York: de Gruyter.

Niebuhr, O. & Michalsky, J. (2018). Virtual reality

simulations as a new tool for practicing presentations

and refining public-speaking skills. In Proceedings of

the 9th International Conference on Speech Prosody

2018, 309-313. International Speech Communication

Association (ISCA).

ERSeGEL 2024 - Workshop on Extended Reality and Serious Games for Education and Learning

692

Notaro, A., Capraro, F., Pesavento, M., Milani, S. & Busà,

M.G. (2021). Effectiveness of VR immersive

applications for public speaking enhancement. In Proc.

IS&T Int’l. Symp. on Electronic Imaging: Image

Quality and System Performance XVIII, 2021, 294-1 -

294-7. doi: 10.2352/ISSN.2470-1173.2021.9.IQSP-294

Radianti, J., Majchrzak, T. A., Fromm, J., & Wohlgenannt,

I. (2020). A systematic review of immersive virtual

reality applications for higher education: Design

elements, lessons learned, and research agenda.

Computers & Education, 147, [103778]. doi:

10.1016/j.compedu.2019.103778

Remacle, A., Bouchard, S., Etienne, A., Rivard, M., &

Morsomme, D. (2021). A virtual classroom can elicit

teachers’ speech characteristics: evidence from acoustic

measurements during in vivo and in virtuo lessons,

compared to a free speech control situation. Virtual

Reality, 25(4), 935-944.

Rohrer, P. L., Vilà-Giménez, I., Florit-Pons, J., Gurrado,

G., Gibert, N. E., Ren, A., Shattuck-Hufnagel, S., &

Prieto, P. (2021, February 24). The MultiModal

MultiDimensional (M3D) labeling system. doi:

10.17605/OSF.IO/ANKDX

Schneider, J., Borner, D., van Rosmalen, P., & Specht, M.

(2017). Presentation Trainer: what experts and

computers can tell about your nonverbal

communication. Journal of Computer Assisted

Learning, 33(2), 164-177. doi: 10.1111/jcal.12175

Selck, K., Albert, T., & Niebuhr, O. (2022). “And miles to

go before…”–Are speech intensity levels adjusted to

VR communication distances?. In Book of Abstracts of

the 13th Nordic Prosody Conference, Sønderborg,

Denmark, 44-46.

Slater, M., Pertaub, D. P., & Steed, A. (1999). Public

speaking in virtual reality: Facing an audience of

avatars. IEEE Computer Graphics and Applications,

19(2), 6-9.

Tamarit, L., Goudbeek, M., & Scherer, K.R. (2008).

Spectral slope measurements in emotionally expressive

speech. In proceedings of SPKD-2008, paper 007.

Thoen, A., Steyaert, J., Alaerts, K., Evers, K., & Van

Damme, T. (2023). A Systematic Review of Self-

Reported Stress Questionnaires in People on the Autism

Spectrum. Review Journal of Autism and

Developmental Disorders, 10(2), 295-318. doi:

10.1007/s40489-021-00293-4

Valls-Ratés, Ï., Niebuhr, O., & Prieto, P. (2021). Effects of

public speaking virtual reality trainings on prosodic and

gestural features. Proc. 1st International Conference on

Tone and Intonation (TAI). University of Southern

Denmark. doi: 10.21437/TAI.2021-44

Valls-Ratés, I., Niebuhr, O., & Prieto, P. (2022). Unguided

virtual-reality training can enhance the oral

presentation skills of high-school students. Frontiers in

Communication, 7, [910952]. doi:

10.3389/fcomm.2022910952

Valls-Ratés, I., Niebuhr, O., & Prieto, P. (2023).

Encouraging participant embodiment during VR-

assisted public speaking training improves

persuasiveness and charisma and reduces anxiety in

secondary school students. Frontiers in Virtual Reality,

4, [1074062]. doi: 10.3389/fvir.2023.1074062

Van Ginkel, S., Ruiz, D., Mononen, A., Karaman, A. C., de

Keijzer, A. & Sitthiworachart, J. (2020). The impact of

computer-mediated immediate feedback on developing

oral presentation skills: An exploratory study in virtual

reality. Journal of Computer Assisted Learning, 36. doi:

10.1111/jcal.12424

Wolpe, J. (1969). The Practice of Behavior Therapy,

Pergamon General Psychology Series, 1. 4th Edition.

Elmsford, NY, US: Pergamon Press.

Xu, Y. (2013). ProsodyPro - A Tool for Large-scale

Systematic Prosody Analysis. In Proceedings of Tools

and Resources for the Analysis of Speech Prosody

(TRASP 2013), Aix-en-Provence, France. 7-10.

VR Public Speaking Simulations Can Make Voices Stronger and More Effortful

693