A preliminary Survey of the Literature on Multilingual, Multimodal Speech Emotion Detection using Bibliometric and Co-Occurrence Analysis

Abstract

Emotion recognition across languages is a rapidly growing field of study. Many different techniques have been developed to determine the emotions of a speaker's words as interest in the study of speech signals has grown. The research that has been done on speech emotion identification makes use of a wide range of methods, including traditional speech analysis and classification strategies, in order to decode the feelings included within signals (SER). This study surveys the field of multilingual and multimodal speech emotion detection. Review topics include speech emotion detection limitations, speech emotion extraction methods, speech emotion databases used, and speech emotion extraction contributions. The purpose of these publications was to evaluate the multilingual and multimodal speech emotion detection in Scopus in the context of research trends and top nations. Based on voice emotion recognition in multiple languages and communication modes, the search syntax. In our essay, we provide an overview of the ten years' worth of study on this subject. The Scopus database was searched using the keywords "voice emotion detection" and "multilingual and multimodal" to find 706 records between January 1, 2013, and October 23, 2022. These records were then assessed using the VOSviewer software.