TY - GEN

T1 - How Children Recognize Emotions from Video and Audio

AU - Ляксо, Елена Евгеньевна

AU - Фролова, Ольга Владимировна

AU - Николаев, Александр Сергеевич

AU - Гречаный, Северин Вячеславович

AU - Филатова, Юлия Олеговна

AU - Ruban, Nersisson

N1 - Conference code: 26

PY - 2024/11/22

Y1 - 2024/11/22

N2 - The paper presents the results of assessing the ability of children from different groups - typical development (TD), autism spectrum disorder (ASD), and intellectual disability (ID) - to recognize the emotions of other people using video and audio tests. The study included 225 children aged 5 to 16 years and 2 adults. We used the original dataset by the Child’s Emotional Development Method (CEDM), selecting from dataset the video recordings of four test tasks performed by children to recognize emotions from photographs of children, photographs of adults, abstract images, and the speech and voice of children (audio test). Children’s ability to recognize emotions correctly was assessed in two ways: by experts’ ratings on a Likert scale and by children’s correct answers. The results showed greater accuracy in recognizing emotional states by TD children compared to children with ASD and ID. Peculiarities of recognition by children with ASD and ID of various emotions using video and audio tests were revealed. The novelty of the results lies in the comparative data on the recognition of emotions by TD children and children with various developmental disorders. The problem of assessing children’s emotion recognition using a standardized approach for further use of data in creating applications that teach children to express and recognize emotions is original and new.

AB - The paper presents the results of assessing the ability of children from different groups - typical development (TD), autism spectrum disorder (ASD), and intellectual disability (ID) - to recognize the emotions of other people using video and audio tests. The study included 225 children aged 5 to 16 years and 2 adults. We used the original dataset by the Child’s Emotional Development Method (CEDM), selecting from dataset the video recordings of four test tasks performed by children to recognize emotions from photographs of children, photographs of adults, abstract images, and the speech and voice of children (audio test). Children’s ability to recognize emotions correctly was assessed in two ways: by experts’ ratings on a Likert scale and by children’s correct answers. The results showed greater accuracy in recognizing emotional states by TD children compared to children with ASD and ID. Peculiarities of recognition by children with ASD and ID of various emotions using video and audio tests were revealed. The novelty of the results lies in the comparative data on the recognition of emotions by TD children and children with various developmental disorders. The problem of assessing children’s emotion recognition using a standardized approach for further use of data in creating applications that teach children to express and recognize emotions is original and new.

KW - Emotions, Recognition, Children with Autism Spectrum Disorders and Intellectual Disabilities, Video and Audio Modalities

KW - Children with Autism Spectrum Disorders and Intellectual Disabilities

KW - Emotions

KW - Recognition

KW - Video and Audio Modalities

UR - https://www.mendeley.com/catalogue/0209bf50-e13f-3bd4-9bac-ce717a59f70f/

U2 - 10.1007/978-3-031-78014-1_11

DO - 10.1007/978-3-031-78014-1_11

M3 - Conference contribution

SN - 978-3-031-78013-4

T3 - Lecture Notes in Computer Science ( Lecture Notes in Artificial Intelligence)

SP - 138

EP - 153

BT - 26th International Conference, SPECOM 2024, Belgrade, Serbia, November 25–28, 2024, Proceedings, Part II

A2 - Karpov, Alexey

A2 - Delić, Vlado

PB - Springer Nature

T2 - 26th International Conference on Speech and Computer

Y2 - 25 November 2024 through 28 November 2024

ER -