JETI

JETI Admin

Abstract

The advancement of Artificial Intelligence has brought about a wider application in different areas ranging from health, n and Entertainment. Music is essential in human’s daily life which is determined by our mood or emotion at a particular moment, a user choice of music is influenced by their mood as well as their past musical tastes and musical substance. Facial expressions are a natural way of conveying emotions, and music is a powerful tool for influencing and reflecting human emotions. Songs are been played based on manual selection of a particular song from the music player or manually playing a particular song from a playlist by manual selection, irrespective of the listener’s mood. However, different songs have different moods they suit which might possibly help in cheering the listener up in case of sad or bad mood. This research focused its attention on the development of a facial emotional-based music player which introduced a powerful music recommendation system that makes music suggestions depending on the user's current mood, which improves user experience and emotional well-being. Convolutional Neural Network with five convolutional blocks, two fully connected layers and an output layer were employed for the detection of facial emotion, FER2013 dataset with 35,887 grayscale photos of facial expression with their related emotional categories like despair, disgust, happy fear, surprise, sad and neutral was used and went through a series of preprocessing techniques like scaling using bicubic interpolation, normalization using Local Binary Pattern (LBP). Filtering using Fast Fourier Transform and label discretization using one-hot encoder. The research employed hold-out evaluation method of 80, 10, 10 training, validation and testing split respectively. The system gave an average accuracy of 75% and best F1 score of 87% on happy mode class. The system outperformed other existing system that used the same dataset. Other architectures of CNN should be employed in the future works. The developed system cannot be incorporated to existing music players for suggestion and this can be done by future researchers.

References

[1] Linnemann, A., Kappes, C., Vorderer, P. and Bente, G. (2015). Music and health communication: A theoretical model and empirical study on the effects of a persuasive message and music on intentions and behavior. Journal of Health Psychology, 20(9), 1155-1166.

[2] Ekman, P (1992). An argument for basic emotions. Cognition & Emotion 6(3), 169-200

[3] Choi, S., Nam, J., Park, K., & Lee, B. (2019). Development of an emotional music player using machine learning. Multimedia Tools and Applications, 78(17), 25017-25030.

[4] Lin, J. Y., Chen, C. H., & Huang, K. C. (2016). A fuzzy music recommendation system based on emotion. Journal of Intelligent and Fuzzy Systems, 30(1), 323-333.

[5] Jin-Young, K., Sang-Hee, C., Hyoung-Gook, K., and Jin-Woo, J. (2014). Affective design of personalized music playlist based on user emotions. Multimedia Tools and Applications, 72(3), 2381-2403. https://doi.org/10.1007/s11042-013-1418-9

[6] Surajit, D., Suman, S. S., and Sivaji, B. (2014). Emotion-based music retrieval using machine learning techniques. International Journal of Computer Science and Mobile Computing, 3(9), 950-956. https://doi.org/10.1111/1755-0998.12700

[7] Kaushik, R., Anand, A., and Bhatia, R. (2016). A music recommendation system based on user emotions and preference. International Journal of Computer Applications, 146(7), 15-21

[8]Yang, C., Yang, J., & Chen, J. (2016). An Affective Music Recommender System based on User Emotional Preferences and Lyrics. Multimedia Tools and Applications, 75(13), 7489-7512. https://doi.org/10.1007/s11042-015-2988-8

[9] Lee, J., Kim, J., & Kim, J. H. (2018). Emotion-Based Music Recommendation Using Brainwave Signals. IEEE Access, 6, 51210-51219. https://doi.org/10.1109/ACCESS.2018.2873659

[10] Abdul, H., Malik, H., and Shaheen, I. (2018). Emotion-Based Music Player Using Facial Expression Recognition. International Journal of Computer Science and Information Security, 16(8), 1-6.

[11] Shreyas, B. K., Prabhu, A. M., and Hegde, A. (2018). Emotion-Based Music Player Using Convolutional Neural Networks. In Proceedings of the 2018 International Conference on Advances in Computing, Communications and Informatics. 303-308..

[12] Amol, S., Prabhu, G., and Suvarna, S. (2019). EEG based emotion recognition for personalized music player. International Journal of Advanced Research in Computer and Communication Engineering, 8(4), 47-52.

 [13] Puri, S., Singh, S., and Dhanjal, S. (2020). Emotion detection using image processing in Python. International Journal of Advanced Science and Technology, 29(1), 1082-1088.

[14] Shaik, S. A., Raju, S. S., and Prasad, G. V. (2020). Real-time facial emotion classification and recognition using deep learning model. International Journal of Innovative Technology and Exploring Engineering, 9(3), 1573-1577.

 [15] So-Yeon Park, Sang-Hyeok Lee, and Chang-Dong Yoo. (2020) "Emotion-based music player using facial expression recognition”. In Proceedings of the 5th International Conference on Multimedia Systems and Signal Processing, 10-13.

[16] Wang, J., Gao, H., Lu, J., & Huang, X. (2021). A Deep Learning-Based Emotional Music Player. IEEE Access, 9, 57555-57562.

PDF