Speech and Opinion Recognition from a Conversation
Tavishi Priyam1, AMJ Muthukumaran2, Himanshu Vinayak3
1Tavishi Priyam*, Department of Computer Science and Engineering, SRM Institute of Science and Technology, Kattankulathur.
2AMJ Muthukumaran, Assistant Professor, Department of Computer Science & Engineering at SRM Institute of Science and Technology, Chennai.
3Himanshu Vinayak, Mobile Developer, Department of Computer Science & Engineering at SRM Institute of Science and Technology, Chennai.
Manuscript received on March 15, 2020. | Revised Manuscript received on March 31, 2020. | Manuscript published on April 10, 2020. | PP: 1189-1193 | Volume-9 Issue-6, April 2020. | Retrieval Number: E2466039520/2020©BEIESP | DOI: 10.35940/ijitee.E2466.049620
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Speech Recognition is an interdisciplinary technique used to convert spoken language into text. It is a sub domain of computational linguistics and can be implemented using Machine Learning and Deep Learning Algorithms. Opinion Mining or Sentiment Analysis is a process which enables identifying opinions expressed by an author in a piece of text computationally. This opinion refers to the polarity of the expressed opinion, i.e. positive or negative. Through this research work, we aim to combine these two natural language processing techniques and devise a system that can take speech as the input and determine the sentiment behind the speakers’ words. The subject of the speech input may vary but the end goal is to recognize whether the attitude of the speaker towards the subject was positive or negative. The input will be converted to text and this text will then be classified using several different machine learning techniques. These include Naïve Bayes’ Classifier, Support Vector Machine, Logistic Regression and Decision Trees. After classification, the results for the three classifiers will be predicted and compared. Future scope of the project includes creating an ensemble of these classifiers to get better accuracy and precision of determining the sentiment of the speaker.
Keywords: Sentiment Analysis, Machine Learning, Natural Language Processing, Opinion Mining, Speech Recognition
Scope of the Article: Machine Learning