• Version
• Download 30
• File Size 242.72 KB
• File Count 1
• Create Date 23 June 2023
• Last Updated 23 June 2023

Audio Sentiment Analysis With Different Approach

Mr.Raju Rathod , Ritesh More, Sharayu Sabde, Vaibhav Zadokar, Rajnandan Vhanabatte

Dept: Electronnics and telecommunication

JSPM’s Rajarshi shahu college of

engineering

Pune, India

Abstract— Over the last few years, sentiment analysis grown significantly. Many strategies are used within academics and industry to automatically categories emotion. The majority of the work done on sentiment analysis over the past few decades has focused on textual sentiment analysis using text mining techniques. But the field of research on audio sentiment analysis is still in its infancy. Here, we've first looked at the methods that different scholars have put forth for sentiment analysis across all data modalities. We then attempted putting some of those strategies into practice. In this essay, we have concentrated on audio sentiment analysis. Researchers keep discovering about audio sentiment analysis, consequently new approaches are being used to examine audio data. Here, we attempt to use machine learning to categories audio into different feelings. We discussed multimode analysis in addition to audio sentiment in this study. Multimodal sentiment analysis includes categorizing emotions using a variety of data types, including text, audio, and video. After combining the individual modality with attention networks for audio sentiment analysis, our technique also produced respectable results. The goal was to create a system that, when data is given into it, can recognize six different emotions in an audio file, including anger, joy, disgust, sorrow, fear, and surprise.

Keywords:( Audio sentiment, Machine learning, RAVDESS, Natural language processing NTP)