A Speech Recognized Dynamic Word Cloud Visualization for Text Summarization

Abstract

A word cloud is a vivid graphic depiction of the words delimited in a section of text engendered by web-oriented tools and it is a straightforward and visually appealing visualization method for text. It has a versatile utility to provide an overview by fractionating text down to those words with higher frequency. Usually, a word cloud can be built by pure text summarization where monotonous data is turned into a stunning visualization format to highlight important textual content and immediately convey crucial information. It is easy to understand visual data rather than any form of data as they are informal to comprehend and memorize. Word clouds can be exercised to analyze data from social network websites and stream service platforms for identifying the trending updates. This approach is an effective technique used to solve text analysis tasks and evaluate it as a qualitative behavior. The input data may consist of articles and pronouns like `a', `the', `and', `he', etc which have to be ignored and the frequency of significant words should be considered to produce a word cloud. Depending on the frequency of the word, the size of the word will vary while producing a word cloud. In addition to the pre-existing static text word cloud, our method proposes an improvement for the dynamic data input from the user, with speech recognition enabled with guaranteed semantic coherence and spatial stability of the words analyzed. Our work is personified in an interactive visual study system that aids users to perform text analysis and derive insights from a huge collection of documents and voice commands.