Speech Emotion Recognition: Sentiment Analysis on Speaker Specific Speech Data

EasyChair Preprint 5592

Abstract

Sentiment analysis has evolved over the past few decades. Most of the work revolves around text sentiment analysis using text mining techniques, but audio sentiment analysis is still in its infancy. Recognizing emotions from speech signals is an important but challenging component of human-computer interaction (HCI). Many techniques have been used in the Speech Emotion Recognition (SER) literature to extract emotions from signals, including many well-established speech analysis and classifications. Techniques. In this proposed model, we perform sentiment analysis for speech discriminated by speaker protocols to capture the emotions of each speaker involved in the conversation. We analyze various techniques for performing speaker discrimination and sentiment analysis to find efficient algorithms to perform this task.In this attempt to extract the human emotions from affective states of speech, this literature uses many different techniques to make machines understand the human emotions. This paper covers the different datasets, extracted emotions & different attempts made towards analysis of human sentiments.

Keyphrases: Convolutional Neural Network, Deep Neural Network, deep learning, speech emotion recognition

@booklet{EasyChair:5592,
  author    = {Aditya Kulgod and Shubham Shinde and Chinmay Phadke},
  title     = {Speech Emotion Recognition: Sentiment Analysis on Speaker Specific Speech Data},
  howpublished = {EasyChair Preprint 5592},
  year      = {EasyChair, 2021}}