Facial Emotion-Based Song Recommender System Using CNN
Ashish Tripathi, Abhijat Mishra, Rajnesh Singh, Bhoopendra Dwivedy, Amit Kumar, Kuldeep Singh · International Journal of Engineering Trends and Technology · 2024
It is observed that many times, people are not able to recognize what kind of song they really want to listen to on the basis of their current mood. Sometimes, people end up searching for the perfect song according to their mood, and it eventually leads to a waste of time in finding the exact requirements of the song. In the era of technology and research, specifically in the world of Artificial Intelligence (AI), implementing these technologies in the advancement of song recommender systems will help people recognize the exact requirements and recommend songs accordingly. It will be the perfect combination of technology and the requirements of the user. This research basically focuses on the recommendation of a song to the person based on his/her current mood. According to the mood of an individual, the song is recommended. The process goes like this: the user first takes a photo of the user with the help of a webcam on the laptop, with the user’s permission. After that, the number of photos is matched with the data stored, and when the particular emotion is identified with the help of a CNN (convolutional neural network), it is then redirected to YouTube. According to the mood of the user, the song is played. Hence, the basic requirement of the song recommender system is completed.