SPEECH EMOTION RECOGNITION USING CNN-LSTM

SPEECH EMOTION RECOGNITION USING CNN-LSTM

Ms Gayathri R ¹, Arun Kumar B², Inbanathan S³, Karthick S⁴

¹Assistant Professor (Sr.Gr), Department of Electronics and Communication Engineering, Sri Ramakrishna Institute of Technology, Coimbatore, Tamil Nadu, India.

²Final Year Student, Electronics and Communication Engineering, Sri Ramakrishna Institute of

Technology, Coimbatore, Tamil Nadu, India. 

³Final Year Student, Electronics and Communication Engineering, Sri Ramakrishna Institute of

Technology, Coimbatore, Tamil Nadu, India. 

⁴Final Year Student, Electronics and Communication Engineering, Sri Ramakrishna Institute of

Technology, Coimbatore, Tamil Nadu, India. 

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract -Speech emotion recognition is a rapidly growing field of research that aims to automatically identify emotions from speech signals. This paper presents a speech emotion recognition using machine learning techniques. The study begins by providing an overview of the various approaches used in speech emotion recognition, including feature extraction, feature selection, and classification. These selected features like pitch, MFCC are compared with the existing datasets in databases. and baased on the features the audios are classified using CNN LSTM algorithm. This model is trained in the free environments like collab using Python, and for User interface and HTML, CSS is used.

Key Words:  Speech Emotion, MelFrequency Cepstral Coefficient, CNN, LSTM