• Version
• Download 12
• File Size 209.30 KB
• File Count 1
• Create Date 28 May 2023
• Last Updated 28 May 2023

Text to Speech Conversion Using Sentiment Analysis

Bhumika Gundale, Harsh Baviskar, Neha Jagtap, Sanskar Mankar

Dept of Information Technology
MIT ADT University

ABSTRACT

In this paper, we propose emotion based text to speech model which converts text to speech in such a way that it takes into account all the emotions of the text and incorporates all the emotions into speech. Although current text-to-speech models are able to generate high-quality speech, generated voice is often not perceptually identifiable by its intended emotion category .To address this problem, we propose a system to synthesize emotional and natural human like sound from unstructured text. The entire system has two parts, first part deals with extracting emotion out of text using classification. The Second part is to modulate the voice to give it an emotional and human like base. Sentiment analysis, also referred to as opinion mining, is an approach to natural language processing (NLP) that identifies the emotional tone behind a body of text. This study is expected to improve the different methodologies in sentiment analysis as well as in generation of synthetic speech.

KEYWORDS

Text to Speech; Sentiment Analysis; Emotion Detection.