DESCU: Dyadic emotional speech corpus and recognition system for Urdu language

Qasim, Muhammad; Habib, Tufail; Urooj, Saba; Mumtaz, Benazir

doi:10.1016/j.specom.2023.02.002

Cited by 4 publications

(1 citation statement)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In today's society, student-centered teaching methods are becoming mainstream, and personalized and contextualized learning are receiving increasing attention [1][2][3][4][5]. Therefore, it is necessary for teachers to understand and adjust students' learning emotions in a timely manner to achieve optimal learning outcomes [6][7][8][9][10][11]. The smart classroom is an important product of new education technology, which uses the latest computer, communication, and Internet technologies to achieve deep perception, intelligent analysis, and accurate delivery of the education process [12][13][14][15][16][17].…”

Section: Introductionmentioning

confidence: 99%

Emotion Recognition in Learning Scenes Supported by Smart Classroom and Its Application

Zhu,

Zheng,

et al. 2023

View full text Add to dashboard Cite

Emotion recognition technology is one of the important applications of artificial intelligence and machine learning in the field of education. By recognizing the emotions of students in learning scenes, teachers can better understand the learning status of students and provide them with personalized learning resources and help. Current emotion recognition methods are mainly based on static facial emotions, neglecting the temporal features of facial emotions, which may lead to inaccurate recognition results. In order to overcome these challenges, this study conducts research on emotion recognition and its application in learning scenes supported by smart classrooms. The Transformer encoder is used to extract the temporal features of student facial emotions based on learning scenes, i.e., the selfattention module of the encoder is used to extract the temporal features of facial emotions in learning scenes. Residual attention networks, Transformers, and non-local neural networks are used to extract facial emotion features from different perspectives and levels. The combination of Vision-Transformer (ViT) and NetVLAD enables the model to learn the features of data from multiple perspectives, thereby improving the generalization ability of the model. The experimental results verify the effectiveness of the constructed model.

show abstract