Emotion Recognition from Speech Using Convolutional Neural Networks

Files

Access Instruction 899.pdf (104.12 KB)

Date

2023

Authors

Journal Title

Journal ISSN

Volume Title

Publisher

Springer Science and Business Media Deutschland GmbH

Abstract

The human voice carries a great deal of useful information. This information can be utilized in various areas such as call centers, security and medicine among many others. This work aims at implementing a speech emotion recognition system that recognizes the speaker’s emotion using a deep learning neural network based on features extracted from audio clips. Different datasets including the RAVDESS, EMO-DB, TESS and an Emirati-based dataset were used to extract features. The features of each dataset were used as the input that would be fed into a convolution deep neural network for emotion classification. Several models were implemented based on extracted features from each dataset. The top three models that produced the best results were reported. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Description

Keywords

CNN, Emotions classification, MFCC, Speech

Citation

Mahfood, B., Elnagar, A. & Kamalov, F. (2023). Emotion Recognition from Speech Using Convolutional Neural Networks. In A. Khanna, Z. Polkowski, & O. Castillo (Eds.) Proceedings of Data Analytics and Management . Lecture Notes in Networks and Systems, 572, (pp. 719 – 731). Springer, Singapore. https://doi.org/10.1007/978-981-19-7615-5_59

URI

https://doi.org/10.1007/978-981-19-7615-5_59
 https://hdl.handle.net/20.500.12519/899

DOI

Collections

Department of Electrical Engineering