Sentiment Analysis across Modalities: A Comprehensive Review of Text and Audio Approaches

Sentiment analysis, an essential component of natural language processing, reveals the complex web of human emotions and viewpoints present in a variety of communication formats. Text-based as well as audio-based sentiment analysis are two areas where this survey that closely examines sentiment analysis. The digital era’s explosion of textual and auditory data has sparked a great deal of research in these fields, producing a wide range of methods and models. Sentiment analysis algorithms have become increasingly complex as a result of the availability of textual data from sources like social media and news articles. Concurrently, there has been a surge in interest in audio-based sentiment analysis because of the increasing use of audio content in podcasts, customer service interactions, and multimedia platforms. This survey provides an in-depth understanding of the recent methods, techniques and approaches in sentiment analysis using text and audio. It carefully describes in detail various developments in these fields, encouraging interdisciplinary idea-sharing. This paper presents a comprehensive overview of sentiment analysis advancements by offering a comprehensive review of research papers, with a focus on relevant models and techniques.