ANALYSIS OF EMOTIONS USING VOICE FEATURES
DOI:
https://doi.org/10.34185/1991-7848.itmm.2025.01.119Keywords:
emotion recognition, emotion recognition through sound, machine learning, emotion analysis, speech processing.Abstract
The paper describes methods for recognizing human emotions. This area of artificial intelligence is developing rapidly and plays a key role in improving the efficiency of human-computer interaction. The study analyzes existing approaches to recognizing emotions using voice features. This paper discusses methods for analyzing emotions based on voice features. The main parameters, such as frequency, tempo, intonation, timbre, and volume, are described, which make it possible to determine the emotional state of a person by his or her speech. Modern approaches are discussed, including the use of machine learning algorithms and neural network models for processing audio files and classifying emotions. This study contributes to the development of emotion research by providing a deeper understanding of human emotional states.
References
Z. Huang, M. Dong, Q. Mao, Y. Zhan. Speech emotion recognition using CNN. Proceedings of the 22nd ACM international conference on multimedia, Association for Computing Machinery, New York, NY, USA (2014), pp. 801-804, DOI: 10.1145/2647868.2654984
E. Lakomkin, C. Weber, S. Magg, S. Wermter. Reusing neural speech representations for auditory emotion recognition. (2018), DOI: 10.48550/ARXIV.1803.11508.
Dmitrieva I.S., Bimalov D.V. Analysis of emotions using facial expressions and voice signs. Regional inter-university collection of scientific works. - Issue 3(158). - Dnipro:
USUST, 2025. p. 21-27. DOI 10.34185/1562-9945-3-158-2025-03
