Anda di halaman 1dari 12

NEURAL NETWORK APPROACHES FOR

CHILDREN’S EMOTION RECOGNITION IN


INTELLIGENT LEARNING APPLICATIONS

Felix Albu
Valahia University of Targoviste, Romania
Daniela Hagiescu, Mihaela Puica
Danimated SRL, Bucharest, Romania
Liviu Vladutu
Polytechnic University of Bucharest, Romania
Outline
Motivation and objectives

Neural network structures

Emotion recognition results

Affect–aware learning applications

Conclusions
Motivation and objectives
The evaluation of school-aged children’s emotion recognition
is a challenging problem.
There is a need for an automatic emotion recognition in order
to keep the child interested in a learning application.

Objectives:
 to investigate different neural network structure
performance on emotion recognition from speech and
images
 To detect the mood of the child
Feature selection

34-element feature vector for a speech file


Feature selection

An example of a feature vector for a “negative” emotion


Speech emotion classification
Speech emotion classification
Speech emotion classification
EMOTION EVALUATION FROM FACE DETECTED IMAGES
The first step is to detect the landmarks on children’s faces: eyes,
eyebrows and mouth.

11 distances are computed: two distances between centre of the left/right


eye and center of the left/right eyebrow, two distances between corner of
the left/right eye and corner of the left/right eyebrow, four distances
measuring width and height of the eyes, distance between the corner of
the eyebrows and width and height of the mouth.
AFFECT-AWARE LEARNING APPLICATIONS
- We consider that the detection of a negative user affective
state should trigger motivation-increase mechanisms.
- Also, the successful completion of a learning cycle should
trigger a reward mechanism. This concept of applying game
mechanics to non-game contexts (in this case, an
educational context) is called gamification, since the reward
mechanisms is always present in the game design through
passing to the next level of the game, or earning extra-points
or lives.
- One possible decision that the tutor can take when the
sentiment-analysis algorithm detects a negative feeling (like
boredom, sadness or frustration) is to offer the child some
kind of hints using easy, cheerful, multimedia-based multiple-
choice quizzes.
Conclusions
Several network structures for an emotion recognition
application were investigated.
The emotion recognition is based on recorded speech signals
and face detected images.
The robustness of the classification rate to noise on both
speech and image signals was investigated.
Our future work will be focused on correlating the results using
images and speech and improve the results by selecting the
most suitable features for the chosen network architecture.
Also, we take into consideration improving our classification
tools.
Please address your questions to:

felix.albu@valahia.ro
danimated.233@gmail.com
mihaela.puica@gmail.com
liviu.vladutu@gmail.com

Anda mungkin juga menyukai