Collaborative Synchronous Hybrid Learning Environments:

Opportunities and Audio/Acoustic Quality Challenges

Héloïse Tudela

, Mathieu Serrurier

, Nader Mechergui

2,3

, Yannick Brudieux

and Mériem Jaïdane

2,3

Quietam Studio, 16 rue de l’Hermite, 33520 Bruges, France

dB.Sense, Innovation Centre of Tunis El Manar University, 2092 Tunis, Tunisia

Signals and Smart Systems Lab., ENIT, Univ. of Tunis el Manar, 2092 Tunis, Tunisia

Keywords: Information Technologies, Collaborative Learning, Synchronous Hybrid Learning, Challenging Acoustic

Environments, Audio Quality.

Abstract: This exploratory study focuses on both pedagogical and technological challenges to anticipate teachers' needs

for information technologies that support a collaborative synchronous hybrid learning environment. This is

based on the OECD's PISA assessment mentioning a decline in students’ academic performances, especially

in problem-solving skills and collaboration, and the growing interest for digital tools dedicated to the learning

environment. Our survey confirmed both the teachers' interest in specific innovative pedagogical scenarios,

made possible by collaborative synchronous hybrid learning environments, and their concern about audio

quality problems, already encountered in classic distance learning situations. This survey is supported by an

"in lab" protocol which explores the technical challenges of audio and acoustic quality associated with two

multi-group scenarios in comparison with the classic scenario. We identified that interfering speech issued

from several speakers who are relatively close to each other and sometimes mobile in the classroom are

creating drastic intelligibility issues that are always more acute for the distant person. The survey revealed

that this form of hybrid learning could lead to specific learning difficulties and fatigue for both students and

teachers, which could be objectively assessed.

1 INTRODUCTION

The recent Covid-19 epidemic and the restrictions

implemented to minimize the spread of the disease

highlighted new challenges, especially in education.

The OECD’s PISA assessment of learning

performances in 2022 revealed an “unprecedented

drop” in scores of assessed countries, showing the

impact of the epidemic on the educational system

(OECD, 2023). More specifically, teachers faced

challenges in the implementation of collaborative

learning, even though those techniques have been

increasingly favoured for its social, academic and

psychological benefits (Laal & Ghodsi, 2012; Marreh

& Velankar, 2024). In 2015, reports from the OECD

already emphasized the importance of developing

students’ “collaborative problem-solving skills and

attitude towards collaboration” (OECD, 2017). The

2022 PISA assessment also highlighted that the

decline in performances began “well before” the

beginning of the pandemic and could indicate a more

systemic issue (OECD, 2023).

To tackle the drop in performances and future

challenges that may occur due to circumstantial

events, companies providing educational resources,

labs and schools need to work together to provide

sustainable tools that support a fundamental

rethinking of teaching and learning.

The European iTEC Innovative Technologies for

Engaging Classrooms project, involving 26 partners

(14 ministries of education, technology providers and

research organisations), was promoting the use of

technology and innovative educational tools to

improve educational performances (Ellis et al., 2015).

This extensive study showed that integrating digital

tools and platforms to the learning environment

improved collaboration and pedagogy for both

students and teachers (Lewin & McNicol, 2015).

Therefore, efforts were made to develop synchronous

hybrid classrooms: students both on-site and remote

could participate and learn simultaneously. Even

512

Tudela, H., Serrurier, M., Mechergui, N., Brudieux, Y. and Jaïdane, M.

Collaborative Synchronous Hybrid Learning Environments: Opportunities and Audio/Acoustic Quality Challenges.

DOI: 10.5220/0013352300003932

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

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 1, pages 512-520

ISBN: 978-989-758-746-7; ISSN: 2184-5026

though synchronous hybrid learning has proved

effective in improving problem-solving skills (Ariani

et al., 2022) as well as students engagement, retention

and communication, a lot of “pedagogical and

technological” challenges persists (Raes et al., 2020).

The growing interest towards collaborative work

mentioned before incentivised teachers to develop its

practice hybrid synchronous settings. However,

studies showed that available video conference tools

that are not intentionally designed for this purpose

were challenging to use for collaborative hybrid

synchronous work (Glazunova et al., 2023; Ortega-

Arranz et al., 2024; Raes et al., 2020).

The orchestration load, the lack of experience

with digital tools and synchronous hybrid learning

environments, the availability of live interactions

between on-site and remote participants are major

areas of improvements for teachers and students when

faced with these new tools (Carruana Martín et al.,

2023; Racheva & Peytcheva-Forsyth, 2024; Raes et

al., 2020). In addition to these pedagogical

challenges, several studies highlighted the key role of

the audio setting to ensure a successful learning

experience (Detyna et al., 2023; Raes et al., 2020).

These observations lead us to investigate the

pedagogical and technological challenges of

collaborative synchronous hybrid learning. After

describing our approach towards this subject, we

designed an exploratory study structured in 2 parts.

The first part consists of a preliminary survey of

middle and high school teachers in France that aimed

at assessing:

- their insights on synchronous hybrid learning

tools, based on their experience of already

available tools,

- their experience of face-to-face collaborative

group work,

- their perspectives and concerns regarding

pedagogical scenarios made possible by

collaborative synchronous hybrid learning.

The second part was designed based on responses

from the survey mentioning technical challenges of

audio and acoustic quality associated with this

learning environment. The design of dedicated

algorithms to enhance the audio experience required

a standardised -adapted ITU-T- subjective assessment

of speech quality (intelligibility, comfort, etc.)

according to different pedagogic scenarios mentioned

in the survey.

The main objective of this exploratory study is to

anticipate technical and pedagogical concerns as soon

as they are identified in order to co-develop solutions

and tools fitted to teachers’ and students’ needs

regarding collaborative hybrid synchronous learning.

2 BACKGROUND AND

APPROACH

It's interesting to note that as early as 2001, a French

study by GRESEC (Groupe de Recherche sur les

Enjeux de la Communication) pointed out that the

failure of collaborations between suppliers of

distance learning solutions and teachers - customers

of these solutions - was due in particular to

(translation) "the conflict between two logics: the

logic of manufacturers, focused on efficiency (optimal

use of technology) and the logic of users, based on

usefulness for users, particularly in relation to the

needs of learners" (Ologeanu, 2001). Our exploratory

study attempts to approach the problems of

collaborative synchronous hybrid learning from these

2 points of view, while being aware of the difficulty

of jointly conceptualising such an environment.

In order to conceptualise the underlying

constraints and challenges of complex pedagogical

scenarios, it is beneficial to be able to have a

“physical” representation of the situation. For this

purpose, the French company Quietam Studio

(Bordeaux, France) inaugurated in 2023 the

ClassLab, as well as the MAIA online platform

dedicated to synchronous hybrid collaborative

activities. Figure 1 shows a hybrid collaborative

tested configuration of ClassLab.

Figure 1: Conceptualisation of ClassLab.

The ClassLab is composed of 3 unconnected table

islands (foreground), an asynchronous broadcast for

a seated group (right) and 3 connected table islands

for synchronous interaction both face-to-face and

remotely in real time (background). It is the analysis

of this synchronous collaborative configuration that

we focus on here. These equipment allow the study of

real-life pedagogical scenarios and the audio/acoustic

challenges that will need to be resolved on the

Collaborative Synchronous Hybrid Learning Environments: Opportunities and Audio/Acoustic Quality Challenges

513

associated digital platforms. The conceptualisation

needs to be both pedagogical and technical.

Pedagogically, involved parties should bear in

mind that the collaborative problem-solving

competency was defined by PISA as “the capacity of

an individual to effectively engage in a process

whereby two or more agents attempt to solve a

problem by sharing the understanding and effort

required to come to a solution and pooling their

knowledge, skills and efforts to reach that solution”.

Its three collaborative problem-solving associated

skills are establishing and maintaining shared

understanding, taking appropriate action to solve the

problem and establishing and maintaining team

organisation. Those skills have to be considered and

developed in a digital context (OECD, 2017),

therefore, the structure of the exploratory survey

carried out is designed to anticipate the pedagogical

opportunities and technical concerns associated with

this new mode of collaboration, based on teachers'

past experience of face-to-face collaborative practice

and synchronous hybrid learning.

Technically, involved parties must be able to

conceptualise the encountered scenario

schematically, then algorithmically and finally

optimise the integration of algorithms dedicated to

enhancing this new hybrid multi-group collaborative

experience on a digital platform. To do this, we need

to draw inspiration from problems encountered in

traditional video conferencing situations, in order to

design and assemble the associated technological

building blocks. State of the Art 2023 on Audio Signal

Processing in the 21st Century: The important

outcomes of the past 25 years provides an accurate

summary of advances and bottlenecks in

multichannel audio (Richard et al., 2023). However,

the challenges resulting from a collaborative

synchronous hybrid learning environment are

seemingly unprecedented. As in the classic hands-

free video conferencing situation, the main difficulty

lies in this time it's multi-loudspeaker/ multi-

microphone coupling from neighbouring groups.

3 SURVEY

Our survey was designed to explore the practices,

challenges, and perceptions of teachers regarding

collaborative learning and distance education, as well

as their potential combination in hybrid modes.

3.1 Method

After gathering basic demographic and professional

information about the participants, we focused on

assessing how teachers integrate digital technologies

into their distance teaching practices in order to better

understand the level of skill and comfort of teachers

with online teaching. We also asked their feedback on

their use of digital tools, especially the quality of their

teaching experience. Teachers were then questioned

about their use and feedback on collaborative

activities. Finally, we introduced the concept of

collaborative hybrid (a combination of face-to-face

and distance) learning environment and gathered their

perspectives and concerns regarding the

implementation of these methods in their classrooms.

The general objective of the survey is to collect

data to analyse how different practices affect teaching

and learning, identify barriers to their effectiveness,

and evaluate the interest in innovative hybrid

approaches. This study aims at identifying

pedagogical scenarios directly from the source - the

teachers - to design technical solutions directly from

those real-life scenarios. This study could give

insights to further improve educational policies and

teaching practice by promoting a more thoughtful

integration of technologies and collaborative methods

in education.

The exploratory survey was distributed to around

100 teachers between June, 2023 and December 2023

and 26 teachers completed the survey. The survey was

distributed to middle schools and highschools in the

Nouvelle Aquitaine region (France), as well as

professional training organisations and members of

the INSPE (Institut National Supérieur du Professorat

et de l'Éducation). The survey was relayed by

emailing and posts via Linkedin. No benefits or

rewards were offered.

13 men and 13 women responded to the survey.

The vast majority of the respondents teach in French

schools, only 2 teach in Canadian schools. More than

42% of the respondents have been teaching for more

than 20 years, around 23% have 10 to 20 years of

experience teaching, 19% have 3 to 10 years of

experience and around 16% have less than 3 years of

experience. The respondents specialise in very

diverse subjects, from languages to sciences,

technology, management and economy. The

respondents were distributed in 3 groups :

approximately 40% teach in middle school, 26%

teach in highschool and 34% teach in other types of

schools such as professional training schools or in

multiple schools.

CSEDU 2025 - 17th International Conference on Computer Supported Education

514

3.2 Results

The first part of the survey inquires about teachers'

usage of collaborative work in the learning

environment. More than ¾ of the respondents use

collaborative work at least once a month, 40% of

which very regularly (at least once every week).

Learners are presented with diverse activities when

doing collaborative work, such as “thematic studies”,

“bibliographical research” or “problem solving”.

Furthermore, several teachers mention that

collaborative work often results in a specific type of

restitution (“oral/exposé”, “creation of a poster”,

“production of videos”). More than half of the

teachers mainly organise groups of 3 to 5 students.

The majority of teachers surveyed believes that

collaborative work is beneficial for students' learning

and personal development: almost 75% of

respondents believe it improves self-confidence, 87%

believe it improves critical thinking and teamwork,

over 80% believe it improves communication

between students and group cohesion, and more than

half think it is a way to motivate students and avoid

school failure. In general, our survey indicates that

teachers favour the use of collaborative work methods

-including hybrid- for student development,

particularly their attitude towards collaboration.

In order to better understand their needs regarding

digital resources, we focused the second part of the

survey on their experience with those devices. Nearly

65% of the surveyed teachers use digital learning

resources regularly in their classroom. Although the

majority of respondents generally report having had

correct audio and video experiences with the tele-

teaching tools, several recurring problems are raised

by users: more than 40% reported having regular

connection and interruption issues, and half of them

experienced regular latencies. In addition, the

majority have noticed echoes or disturbing and

unwanted noises during video conferences, of which

a quarter regularly. Most mention “ambient noise”

such as an “urban atmosphere” or “keyboard noises”,

interfering words and two responses mention Larsen

effects between several microphones and speakers.

These problems related to the use of current video

conferencing tools generate an extra effort to

understand important messages for almost 42% of

respondents and more than 65% report that their

students regularly raise these concerns during use.

These results reveal that the integration of digital

resources in education raises technical problems,

especially audio quality issues, which are not yet

solved by existing solutions.

Lastly, we wanted to gather the opinion of

teachers regarding collaborative hybrid learning, for

which no software has been specifically developed, to

the extent of our knowledge. Nearly 72% of the

surveyed teachers would like to be able to have hybrid

groups in their classroom if they could, namely

groups composed of remote and on-site students.

Together with the ability to better accommodate

students unable to come to class, they also mention

the ability to create mixed classrooms between

regions or countries or to enable teachers of exotic

subjects to teach several classes, improving the

availability of teachers. However, the teachers that

responded that they would not like this setup mainly

explain that they would struggle managing the remote

students. Only 36% of the surveyed teachers had the

opportunity to test this setup, mainly during the

Covid-19 pandemic. However, 50% of the surveyed

teachers think that these setups could improve

student’s problem-solving skills and nearly 75% think

that it can improve collaboration between students,

while allowing teachers to be more available and

enabling a better integration of remote students in the

class.

3.3 Discussion

Our survey showed that teachers are interested in

developing and acknowledge the benefits of

collaborative learning activities in their class, as

shown in similar studies (Laal & Ghodsi, 2012;

Marreh & Velankar, 2024). It also showed the

growing usage of digital resources in the learning

environment, even though teachers and students still

face technical difficulties when using those resources

(Rafiq et al., 2024).

Interestingly, even though very few teachers

already tried using collaborative hybrid or “blended”

learning environments, their responses show a

positive perspective on these scenarios. These

improvements to group work organisation and

performances were already noted as early as 2006

(Graham, 2006). It could be interesting to confront

these perceptions with students’ perspectives and

measure their performances in an hybrid

collaborative setting.

Another interesting result came from confronting

teachers' experiences of collaborative work with their

concerns regarding hybrid collaborative settings:

only 15% of the surveyed teaches experience

disciplinary issues when setting up collaborative

work, but this number raises to 50% when they are

asked if, in their opinion, setting up an hybrid

collaborative environment could induce disciplinary

Collaborative Synchronous Hybrid Learning Environments: Opportunities and Audio/Acoustic Quality Challenges

515

issues. Moreover, nearly 75% of them think that

setting up this kind of environment could drastically

increase their workload. These results showed us that

hybrid collaborative environments raise both

technical and pedagogical concerns. Those concerns

should be addressed together with involved parties

such as teachers, students, educational facilities but

also companies in order to co-create future tools

adapted to hybrid collaborative learning.

In the survey, teachers identified 3 main coupled

issues when they used video conferencing tools that

could also be raised in a hybrid collaborative setup:

echo, howling (Larsen effect) and both ambient noise

and noise from competing speech sources from

neighbour groups. Thus, solutions to these issues

must be developed in order to adapt digital tools to

their use in blended learning environments:

- Multi-input/multi-output echo cancellation in

neighbouring ilot situations, including teacher

mobility, which induces non-stationarity of the echo

channel. The diversity of echo situations in hands-

free sound systems and their resolution in challenging

contexts is of great economic importance and the

subject of international R&D challenges (Shetu et al.,

2024). The current trend is towards hybrid signal/IA

solutions (see for example (Chen et al., 2021)).

- Early detection and reduction of howling in adjacent

ilot situations. This feedback problem is still under

investigation (Zhang et al., 2023).

- Noise reduction in multi-group situations, where the

ambient noise is made up of ambient noise and

interfering -or concurrent- speech from several

speakers who are relatively close to each other and

sometimes mobile in the classroom. Personalised

speech enhancement methods as speaker diarization

("who speaks and when") are algorithmic solution

paths (Eskimez et al., 2021; Wang et al., 2019).

Following this survey, we decided to focus on

investigating the technical challenges raised by a

hybrid collaborative setting on the audio quality of

digital tools.

A subjective audio and acoustic quality evaluation

protocol was set up in a simplified situation with 2

face-to-face groups, each with its own remote

counterpart. The measurement protocol followed ISO

and ITU-T recommendations (P.800, P.808, P.831) as

far as possible in this unusual situation. In the present

study, we will only show the subjective results

regarding the competing speech issue.

Overall, we noticed that between 20% to 54% of

teachers chose the third level out of five when

responding to questions targeting the potential

https://www.itu.int/

advantages of a collaborative hybrid learning

environment (versus 8% to 25% when asked about

collaborative activities only - without the hybrid

setup). We interpreted those results as a struggle to

consider the perspectives and concerns relative to this

kind of learning environment ; emphasising the need

to test real-life scenarios.

4 IMPACT OF HYBRID AND

COLLABORATIVE SCENARIO

ON AUDIO QUALITY

The survey highlighted the impact of the hybrid

situation on audio quality, such as the appearance of

echoes, feedback and environmental noise, which

disrupts intelligibility. To quantify these effects, in

collaborative situation we designed a test protocol

that compares 3 scenarios:

- Scenario 1 classic “L1-D1”: a “one to one” hands

free audio communication between a local room and

a distant room.

- Scenario 2 multichannel “L2-D2”: with two

simultaneous “one to one” hands free audio

communication in separate sessions but in the same

local room.

- Scenario 3 multichannel “L2-D1”: where there is a

“one to two” hands free audio communication

between 2 two persons in the local room with

different devices and 1 person in a distant room.

4.1 Method

The protocol was partially based on the ITU-T

standards

such as P831 (1998) standard on

Subjective performance evaluation of network echo

canceller, P.832 (2000) on Subjective evaluation of

hands-free terminal performance, and adapted to

account for our specific hybrid application.

The aim of this study is to focus on the variation

induced by the two multi-group scenarios in

comparison with the classic scenario. We aimed at

understanding the impact of 5 varying parameters that

could influence the audio quality. The parameters are

presented in Table 1.

CSEDU 2025 - 17th International Conference on Computer Supported Education

516

Table 1: Tested parameters for hybrid communication

protocol.

Parameters Definition 2 levels of

variation

Echo path

Variation of echo

path: simulation

of a teacher

moving between

groups or steady

in a position

V1: steady position

V2: path with

continuous

variation (from

point A to B)

Communication

system distance

Distance between

communication

systems specific

to L2-D2 and L2-

D1 scenarios

V1: distance 1

V2 : distance 2

Speaker voice

level

Speaker voice

level translated

into: low voice,

normal voice,

forced voice

V1: normal voice

V2 : forced voice

Type of

communication

Monologue or

conversational

situation

V1: single remote

speech

V2 : double remote

speech

Multi-group

communication

in local

Interfering

speech in the

classroom

V1: single local

speech

V2: local speech

and interfering

speech from

nei

hbourin

rou

The number of tests for each scenario was:

- L1-D1 classic scenario : 8 test conditions

- L2-D2 and L2-D1 multi-channel scenarios: initially

32 (2^5) test conditions each, but we reduce it to 16,

(2^((5-1) )) each, using a fractional factorial design

approach to estimate the main effects.

For each test condition, 1 to 4 sentences

(phonetically balanced) were used by test

participants. These lists of sentences were taken from

the Combescure and Fournier sentence lists, as well

as the phonetically-balanced HINT test sentences

used in subjective testing, particularly in vocal

audiometry by audiologists (see for example (Chevret

et al., 2012)). In total 310 sentences were used and

mixed so that test subjects only hear the sentences

once in the same scenario.

Finally, test subjects were asked to answer 8

questions Q.i (in French) related to communication

quality for each condition test using ITU-T P.835

standard for “subjective test methodology for

evaluating speech communication systems that

include noise suppression algorithms”:

Q.1: What is your opinion of the connection you have

just used? [5] Excellent [4] Good [3] Fair [2] Poor [1]

Bad

Q.2: Did you experience any difficulties speaking or

listening in this connection? [5] Very strong [4]

Important [3] Medium [2] Small [1] None

Q.3: How would you rate the quality of

communication?

[0] Inacceptable [1] Acceptable

Q.4: How would you rate the degradation caused by

the echo of your own voice?

[5] Imperceptible [4] Perceptible, but not annoying

[3] Slightly annoying [2] Annoying [1] Very annoying

Q.5: How would you rate the other degradations

(truncation, various noises, etc.)?

[5] Imperceptible [4] Perceptible, but not annoying

[3] Slightly annoying [2] Annoying [1] Very annoying

Q.6: How do you find the voice of the person you are

speaking to? [0] Not natural [1] Natural

Q.7: How would you rate the overall sound quality of

the communication?

[5] Excellent [4] Good [3] Fair [2] Poor [1] Bad

Q.8: How would you rate the listening effort required

to understand the message? [5] Absolute relaxation:

no effort [4] Attention required; no appreciable effort

[3] Moderate effort [2] Considerable effort [1]

Meaning incomprehensible despite every possible

effort

The scenario L1-D1 was repeated 3 times with

different persons (6 test subjects in total) and L2-D2

and L2-D1 scenarios were performed once. In total 10

naive test subjects participated in the protocol and all

have signed a consent agreement for using test results

for research purposes.

4.2 Results and Discussion

Firstly, we compared the quality perceived by people

in the different rooms: local and distant. Table 2

shows the average subjective scores on all the test

conditions for each question, Q1 to Q8 for each

scenario (more details are available in (Brudieux et

al., 2024)).

Table 2 shows that the overall quality deteriorates

in L2-D2 and L1-D1 scenarios compared to the

classic L1-D1 scenario, with unacceptable

communication reported in the L2-D1 situation. For

example, the overall sound quality of the

communication decreased from excellent for L1-D1

to fair and poor for respectively L2-D2 and L2-D1 in

the local room.

Collaborative Synchronous Hybrid Learning Environments: Opportunities and Audio/Acoustic Quality Challenges

517

Table 2: Subjective evaluation of audio communication

quality by test subjects in local and distant room for the

scenarios L1-D1, L2-D2 and L2-D1.

Num

Subjective Evaluation

in distant room

Subjective Evaluation

in local room

L1-

L2-

L1-

L2-

Q.1 4.25 4.06 2.92 5 4.43 3.17

Q.2 1.62 2.28 2.85 1.16 1.59 3.71

Q.3 0.91 0.71 0.71 1 0.81 0.32

Q.4 4.87 4.90 4.57 5 5 2.85

Q.5 4.37 3.21 3.07 4.83 3.63 2

Q.6 0.87 0.87 1 1 1 0.5

Q.7 3.91 3.09 3.14 5 3.78 2

Q.8 4.12 3.03 3.28 4.75 4.09 2.03

We further analysed the effect of interfering speech

by specifically extracting subjective quality measures

from the protocol to compare the situation with and

without interfering speech for the 3 questions Q2, Q7

and Q8 in both the local and distant rooms as shown

in figure 2.

Figure 2: Subjective evaluation of audio communication

quality in local and distant rooms for the scenario L2-D2

with/without interfering speech.

The overall results were consistent with a degradation

of the audio quality by interfering speech as measured

in:

(Q2) Speaking and listening difficulties are greater,

(Q7) The overall sound quality of communication is

poorer,

(Q8) The listening effort required to understand the

message is greater.

Interestingly, the degradation of the audio quality

and intelligibility seemed to always be more acute for

the distant person. This can be explained by the fact

that the communication arriving at the remote user is

in mono compared to the local speakers: the signal is

composed of a mixture of all interfering and useful

speech signals, and the user will therefore not be able

to use spatial cues to differentiate the two speech

streams. This result is especially significant, as it

indicates the need for the development of a specific

noise reduction system inside digital tools to reduce

the impact of interfering speech.

Furthermore, in the in-lab situation tested, there

was only one interfering speech stream (from the only

neighbouring group). According to the survey, in

face-to-face group work, teachers usually work with

either groups of 2 students (32%), or groups of

between 3 and 5 students (42%). These issues are thus

likely to be worse in those setups, or setups with more

than two groups, compared to the simplified version

tested in this study. Studies and challenges have

already identified the difficulties of untangling

speech signals and are focused on solving this

specific issue (Cutler et al., 2021).

5 CONCLUSION

This study on anticipating teachers' needs for

Information Technologies to support a collaborative

synchronous hybrid learning environment confirmed:

- teachers' interest in specific innovative pedagogical

scenarios made possible,

- their concern about the extra preparation and

handling workload.

Specifically, the technical concerns regarding the

audio quality were confirmed by subjective in-lab

protocols in simplified situations. Those were

identified to be due to the specific loudspeaker/

microphone acoustic couplings between groups

inherent to the collaborative environment. There is a

pressing need to focus algorithmically on the drastic

audio quality degradation caused by interfering

speech from neighbouring groups, which disturbs

distant speakers in a given group (for more details,

see (Brudieux et al., 2024)).

Verdel (translation) stated that “there is a growing

awareness of the importance of learning spaces,

combining a desire for modularity and diversified use

of digital tools” (Verdel, 2023) thus encouraging the

early identification of technical obstacles that may

hinder the development of innovative digital tools

dedicated to the learning environment.

Our study could benefit from gathering more

teachers’ and students’ feedback, investigating further

the pedagogical scenario “in vivo” and test the

robustness of a hybrid passive/algorithmic solutions

in different class configurations: table arrangement in

islands, use passive insulation panels, number of

groups and number of students per group, ...

Innovative technologies exist to solve some of the

problems encountered in audio quality, such as the

creation of speech zones with self-distributed

CSEDU 2025 - 17th International Conference on Computer Supported Education

518

acoustic swarms (Itani et al., 2023). However, the

scenario of these studies does not correspond to low-

cost algorithmic solutions for democratisation of

collaborative synchronous hybrid learning in schools.

These hybrid collaborative educational scenarios

generate algorithmic solutions that must

automatically adapt to the classroom configuration.

We can reduce the complexity of algorithms, while

improving their robustness, through common sense

solutions: recommendations for modular passive

isolation of groups and classroom behaviour charter,

as specified in the standards for acoustic quality and

comfort in open office spaces

We are aware that the next step is to provide more

theoretical contributions relating to the impact of

these news educational scenarios implemented.

Our survey also highlighted the concern for an

increased fatigue caused by noisy learning

environments and extensive digital tools usage:

nearly 75% of teachers agreed that poor audio quality

combined with the ambient sound inherent to the

collaborative activities could contribute to students

fatigue and lack of involvement. For both

pedagogical and technical reasons, it is of significant

importance that companies and researchers continue

to quantify the effects of collaborative synchronous

hybrid learning on the potential fatigue of students.

This increased mental load has been raised by the

French Senate

and has been objectified on adults by

EEG measurements carried out by Microsoft on the

Microsoft Teams software

. The ZEF scale (Zoom

Exhaustion & Fatigue Scale), validated on a large

scale with adults, seems promising to evaluate the

fatigue of students (Fauville et al., 2021). The scale

was designed to measure the exhaustion and fatigue

felt by users of videoconferencing, caused by intense

visual and close-up contact, reduced mobility, a

higher cognitive and emotional load than in face-to-

face situations, and vocal fatigue. The auditory

fatigue could be integrated into the ZEF scale and

tested on a population of learners.

Altogether, this study emphasises the need for the

development of tools dedicated to hybrid

collaborative learning environments, co-developed

by teachers, educational facilities, software

companies and equipment providers.

ISO 22955:2021 Acoustic quality of open office spaces.

www.senat.fr/rap/r21-259/r21-2591.pdf

ACKNOWLEDGEMENTS

We would like to thank all the participants who

contributed to the creation of the questionnaire, the

animation of the experimental sessions and the

creation of the educational test sequences: Stéphane

Brunel, PhD AdT at INSPE, Lilian Champroy,

Charline Waliyar, teacher at Michel Ragon middle

school (Montaigu, France), and Pascal Rey one of the

teachers of the final class from the Vaclav Havel High

School (Bègles, France).

CONFLICTS OF INTEREST

HT, MS and YB report personal fees from Quietam

Studio as employees, during the conduct of the study.

REFERENCES

Ariani, D., Sumantri, S., Wibowo, F., & Batubara, H. (2022,

mai 8). A Study Examining The Impact of Combining A

Synchronous Hybrid Learning Strategy With Inquiry-

based Learning on Mathematics Problem Solving. The

3rd International Seminar on Education and Human

Technology (ISEHT) 2022, Semarang.

Brudieux, Y., Mechergui, N., Serrurier, M., & Jaïdane, M.

(2024). Verrous technologiques : Qualité audio et

acoustique en mode collaboratif hybride multiflux

[Internal report]. Quietam R&D Department.

Carruana Martín, A., Alario-Hoyos, C., Martínez-Monés,

A., & Delgado Kloos, C. (2023). Analysis of

Orchestration Load and Teacher Agency in Smart

Synchronous Hybrid Learning Environment.

https://doi.org/10.20944/preprints202307.0441.v1

Chen, H., Chen, G., Chen, K., & Lu, J. (2021). Nonlinear

residual echo suppression based on dual-stream

DPRNN. EURASIP Journal on Audio, Speech, and

Music Processing, 2021(1), 35.

https://doi.org/10.1186/s13636-021-00221-8

Chevret, P., Ebissou, A., & Parizet, E. (2012). Evaluation of

speech intelligibility in open-plan offices. 41

International Congress and Exposition on Noise

Control Engineering 2012, INTER-NOISE 2012, 10.

Cutler, R., Saabas, A., Parnamaa, T., Loide, M., Sootla, S.,

Purin, M., Gamper, H., Braun, S., Sorensen, K.,

Aichner, R., & Srinivasan, S. (2021). INTERSPEECH

2021 Acoustic Echo Cancellation Challenge. 4748-

4752. https://doi.org/10.21437/Interspeech.2021-1870

Detyna, M., Sanchez-Pizani, R., Giampietro, V., Dommett,

E. J., & Dyer, K. (2023). Hybrid flexible (HyFlex)

teaching and learning : Climbing the mountain of

implementation challenges for synchronous online and

www.microsoft.com/en-us/worklab/work-trend-index/

Collaborative Synchronous Hybrid Learning Environments: Opportunities and Audio/Acoustic Quality Challenges

519

face-to-face seminars during a pandemic. Learning

Environments Research, 26(1), 145-159.

https://doi.org/10.1007/s10984-022-09408-y

Ellis, W. J. R., Blamire, R., & Van Assche, F. (2015).

Innovative Technologies for an Engaging Classroom

(iTEC). In F. Van Assche, L. Anido, D. Griffiths, C.

Lewin, & S. McNicol (Éds.), Re-engineering the

Uptake of ICT in Schools (p. 1-15). Springer

International Publishing. https://doi.org/10.1007/978-

3-319-19366-3_1

Eskimez, S. E., Yoshioka, T., Wang, H., Wang, X., Chen, Z.,

& Huang, X. (2021). Personalized Speech

Enhancement : New Models and Comprehensive

Evaluation (arXiv:2110.09625). arXiv.

https://doi.org/10.48550/arXiv.2110.09625

Fauville, G., Luo, M., Queiroz, A. C. M., Bailenson, J. N.,

& Hancock, J. (2021). Zoom Exhaustion & Fatigue

Scale. Computers in Human Behavior Reports, 4,

100119. https://doi.org/10.1016/j.chbr.2021.100119

Glazunova, O., Korolchuk, V., Voloshyna, T., & Saiapina,

T. (2023). Evaluation of Microsoft Teams as a Tool for

synchronous Interaction during Distance and Hybrid

Learning. Open Educational E-Environment of Modern

University, 1-11. https://doi.org/10.28925/2414-

0325.2023.141

Graham, C. (2006). Blended learning systems : Definition,

current trends, and future directions (p. 3-21).

Laal, M., & Ghodsi, S. M. (2012). Benefits of collaborative

learning. Procedia - Social and Behavioral Sciences,

31, 486-490.

https://doi.org/10.1016/j.sbspro.2011.12.091

Lewin, C., & McNicol, S. (2015). The Impact and Potential

of iTEC : Evidence from Large-Scale Validation in

School Classrooms. In F. Van Assche, L. Anido, D.

Griffiths, C. Lewin, & S. McNicol (Éds.), Re-

engineering the Uptake of ICT in Schools (p. 163-186).

Springer International Publishing.

https://doi.org/10.1007/978-3-319-19366-3_9

Marreh, S., & Velankar, Y. P. (2024). Recent trends in

collaborative learning : A systematic review and

analysis. International Journal of Evaluation and

Research in Education (IJERE), 13(4), Article 4.

https://doi.org/10.11591/ijere.v13i4.28501

OECD. (2017). PISA 2015 Results (Volume V) :

Collaborative Problem Solving. Organisation for

Economic Co-operation and Development.

https://www.oecd-ilibrary.org/education/pisa-2015-

results-volume-v_9789264285521-en

OECD. (2023). The state of learning and equity in

education in 2022. Organisation for Economic Co-

operation and Development. https://www.oecd-

ilibrary.org/content/component/799d7d80-en

Ologeanu, R. (2001). Visioconférence dans l’enseignement

supérieur : Expérimentations et usages. Les Enjeux de

l’information et de la communication, 2001, 70.

https://doi.org/10.3917/enic.001.0071

Ortega-Arranz, A., Amarasinghe, I., Martínez-Monés, A.,

Asensio-Pérez, J. I., Dimitriadis, Y., Corrales-

Astorgano, M., & Hernández-Leo, D. (2024).

Collaborative activities in hybrid learning

environments : Exploring teacher orchestration load

and students’ perceptions.

Computers & Education,

219, 105105.

https://doi.org/10.1016/j.compedu.2024.105105

Racheva, V., & Peytcheva-Forsyth, R. (2024). Navigating

the Future : Exploring Synchronous Hybrid Learning in

Higher Education Post-Covid (p. 6345).

https://doi.org/10.21125/inted.2024.1661

Raes, A., Detienne, L., Windey, I., & Depaepe, F. (2020). A

systematic literature review on synchronous hybrid

learning : Gaps identified. Learning Environments

Research, 23. https://doi.org/10.1007/s10984-019-

09303-z

Rafiq, S., Iqbal, S., & Afzal, D. (2024). The Impact of

Digital Tools and Online Learning Platforms on Higher

Education Learning Outcomes.

Richard, G., Smaragdis, P., Gannot, S., Naylor, P. A.,

Makino, S., Kellermann, W., & Sugiyama, A. (2023).

Audio Signal Processing in the 21st Century : The

important outcomes of the past 25 years. IEEE Signal

Processing Magazine, 40(5), 12-26. IEEE Signal

Processing Magazine.

https://doi.org/10.1109/MSP.2023.3276171

Shetu, S. S., Desiraju, N. K., Aponte, J. M. M., Habets, E.

A. P., & Mabande, E. (2024). A Hybrid Approach for

Low-Complexity Joint Acoustic Echo and Noise

Reduction (arXiv:2408.15746). arXiv.

http://arxiv.org/abs/2408.15746

Verdel, H. (2023). Interactions entre environnements

numériques et environnements physiques dédiés aux

apprentissages [Doctoral thesis, Université de

Lorraine]. https://theses.fr/2023LORR0355

Wang, Q., Sheth, Y., Moreno, I. L., & Wan, L. (2019).

Speaker diarization using an end-to-end model (World

Intellectual Property Organization Brevet

WO2019209569A1).

https://patents.google.com/patent/WO2019209569A1/

Zhang, H., Yu, M., & Yu, D. (2023). Deep AHS : A Deep

Learning Approach to Acoustic Howling Suppression.

https://doi.org/10.48550/arXiv.2302.09252

CSEDU 2025 - 17th International Conference on Computer Supported Education

520