2021 IEEE 18th India Council International Conference (INDICON),
Journal Year:
2023,
Volume and Issue:
unknown, P. 161 - 166
Published: Dec. 14, 2023
Employing
a
computer
for
automatic
speech-emotion
identification
is
formidable
and
intricate
undertaking.
Speech
emotion
recognition
(SER)
has
gained
significant
popularity
among
academics
over
three
decades
due
to
its
wide
range
of
applications
in
many
industries,
such
as
medical
treatment,
marketing,
customer
service,
driving,
internet
searching,
education.
Researchers
used
approaches
enhance
the
efficiency
categorization.
In
our
work,
we
images
mel
frequency
cepstral
coefficient
(MFCC),
mel-spectrogram,
combination
both
feature
input
2D
convolutional
neural
network
(2D-CNN)
classifier
classify
emotion.
We
trained
model
with
individuals
proposed
Based
on
experimental
results,
observed
that
suggested
MFCC
mel-spectrogram
performed
superior
individual
terms
speech
signal
recognition.
To
assess
efficacy
features,
datasets:
TESS,
RAVDESS,
EMO-DB.
For
EMO-DB,
RAVDESS
datasets,
found
accuracy
categorization
was
88.89%,
100%,
81.2%,
respectively.
Current Topics in Medicinal Chemistry,
Journal Year:
2024,
Volume and Issue:
24(8), P. 737 - 753
Published: Feb. 6, 2024
Background:
SARS-CoV-2,
the
unique
coronavirus
that
causes
COVID-19,
has
wreaked
damage
around
globe,
with
victims
displaying
a
wide
range
of
difficulties
have
encouraged
medical
professionals
to
look
for
innovative
technical
solutions
and
therapeutic
approaches.
Artificial
intelligence-based
methods
contributed
significant
part
in
tackling
complicated
issues,
some
institutions
been
quick
embrace
tailor
these
response
COVID-19
pandemic's
obstacles.
Here,
this
review
article,
we
covered
few
DL
techniques
detection
diagnosis,
as
well
ML
identification,
severity
classification,
vaccine
drug
development,
mortality
rate
prediction,
contact
tracing,
risk
assessment,
public
distancing.
This
illustrates
overall
impact
AI/ML
tools
on
managing
outbreak.
Purpose:
The
focus
research
was
undertake
thorough
evaluation
literature
Intelligence
(AI)
complete
efficient
solution
battle
against
epidemic
domains
diagnostics
disease,
prediction
development.
Methods:
A
comprehensive
exploration
PubMed,
Web
Science,
Science
Direct
conducted
using
PRISMA
(Preferred
Reporting
Items
Systematic
Reviews
Meta-Analysis)
regulations
find
all
possibly
suitable
papers
made
publicly
available
between
December
1,
2019,
August
2023.
along
AI-specific
words,
used
create
query
syntax.
Results:
During
period
by
search
strategy,
961
articles
were
published
released
online.
Out
these,
total
135
chosen
additional
investigation.
Mortality
early
lastly,
incorporation
AI
supervising
controlling
pandemic
four
main
topics
focused
entirely
applications
tackle
crisis.
135,
60
diagnosis
pandemic.
Next,
19
studies
applied
machine-learning
approach
prediction.
Another
22
publications
emphasized
Finally,
remaining
concentrated
applying
AI-based
it.
Conclusion:
We
compiled
from
methodologies
impart
insights
into
various
study.
Our
results
suggest
crucial
characteristics,
data
types,
can
aid
translational
facilitation.
