Results of a Novel Music Player using Speech and Text Emotion Recognition for Mood Uplift

  • Shrikala DESHMUKH, Preeti GUPTA
Keywords: Speech emotion recognition, Text emotion recognition, Probabilistic Neural Network, Support Vector Machine, Positive and Negative Affect Schedule mood scale

Abstract

Music provides a powerful means of conveying and evoking feelings, and it has been shown that emotional regulation is the primary reason for listening to music. With the emergence of digital music in the recent years, the internet has become a major source of retrieving music. However, with abundance of options available sometimes it is frustrating for the music lover to play songs based on his current state of mood. Our work over here implements music player based on emotion recognition which uplifts mood of user. It is a combination of speech emotion recognition and text emotion recognition for emotion recognition. The study takes into consideration five important human emotions namely Happy, Sad, Angry, Fear and Bored. For speech emotion recognition, we used Probabilistic Neural Network (PNN) with Emo Database and for text emotion recognition, we used Natural language processing (NLP) with Support vector machine (SVM) classifier. For speech emotion recognition, pitch and frequency features are used to find emotion behind speech. For this work, the working database of 500 hindi songs has been created out of 700 songs using Positive and Negative Affect Schedule (PANAS) mood scale. We have applied algorithms such as GMM (Gaussian mixture model), HMM (Hidden Markov model), CNN (Convolutional neural network) and RNN (Recurrent Neural network) to our proposed system. It is revealed that our proposed system performs well and shows remarkable accuracy. It shows accuracy of 94.56% with speech emotion recognition using PNN and of 95.76% with combination of speech emotion recognition and text emotion recognition using PNN and SVM accordingly.

Published
2021-08-03
How to Cite
Preeti GUPTA, S. D. (2021). Results of a Novel Music Player using Speech and Text Emotion Recognition for Mood Uplift. Design Engineering, 6222-6232. Retrieved from http://thedesignengineering.com/index.php/DE/article/view/3122
Section
Articles