ANALYZING EMOTIONAL TONES IN VIRTUAL ASSISTANT CONVERSATIONS
J. Venkata Harini1,G. Venu1,G. Vijay Kiran Reddy1, B. Vinayaka Datta1, M. Vinesh Goud1,
Dr Thayyaba Khatoon Mohammed1 and Dr Sujit Das
School of Engineering, Department of AI & ML, Malla Reddy University, Hyderabad – 500043, India.
A B S T R A C T
Emotions play a critical role in human mental life, serving as a primary medium through which individuals express their perspectives and mental states. These emotions can manifest in various forms, such as vocal tones, written text, and facial expressions. This project focuses on developing an advanced system to detect and interpret emotions conveyed through speech. This system, known as the Speech Emotion Analyzer (SEA), aims to identify the emotional state of a speaker based on audio samples. Speech Emotion Analyzer (SEA) can be defined as extraction of the emotional state of the speaker from his or her speech signal through audio.
There are few universal emotions- including Sad, Surprised, Joyfully, Euphoric in which any intelligent system with finite computational resources can be trained to identify or synthesize as required. Emotion detection in audio is essentially a content-based classification challenge that has concepts of natural language processing. Speech emotion analyzer is a classification problem where an input sample (audio) needs to be classified into a few predefined emotions. The SEA operates by extracting and analyzing the emotional content from speech signals. The core challenge lies in classifying these emotional states from audio input using computational techniques. This involves feature extraction from the speech signal—capturing characteristics such as pitch, tone, and tempo—and applying machine learning algorithms to classify these features into predefined emotional categories. The system employs natural language processing (NLP) techniques to enhance emotion recognition by interpreting the context and meaning of the speech.
Keywords:
Speech emotion
Emotion Recognition
Natural Language Processing
Audio Analysis
Machine Learning
Sentiment Analysis
