Recognition of emotional speech using MFCC and Machine Learning Technique

Speech to text conversion and interpreting a person's emotions from their speech are essential tasks in human-computer interaction. Applications such as affective computing necessitate comprehension and reaction to human emotions. We are employing a spectral feature in this project. Technology for determining the various characteristics, such as loudness, tone, and intensity, etc through Mel Frequency Cepstral Coefficients (MFCC).Our primary goal is to categorize eight emotions into a collection of pre-established categories, including neutral, peaceful, pleased, sad, angry, scared, disgusted, and surprised. The suggested technique not only helps to increase voice recognition systems' accuracy but also emphasizes how crucial it is to use coefficients as potent features in this situation. This method has the potential to enhance applications in recommender systems, affective computing, and interaction between humans and computers that require accurate identification of emotions.