Title :
A Proposed Approach with Analysis of Speech Signals for Sentiment Detection
Author :
Atul Tyagi;Nidhi Chandra
Author_Institution :
Comput. Sci. &
fDate :
4/1/2015 12:00:00 AM
Abstract :
In last few decades Sentiment Analysis is one of the most researched area in the field of Speech Recognition. Different methods like use of textual data, face expressions, voice signals, body movements and physiological signals were adopted for sentiment detection. Presented paper discuss about emotions/Sentiment and proposed an approach for analyzing these sentiment based on speech signals of an individual. Paper proposed an approach for extracting speech signals produced by an individual and also provides some results which show variation in pitch and time of an individual´s voice during different emotions. Signals were produced by different individual in different sentiment/moods. Signals are being analyzed using different tools which helps in generating graphs for different signals. This paper discuss about three basic sentiment/emotions were observed in this paper i.e. Angry, sad and neutral (relaxed). The data collected from every individual i.e. From both male and female with different age range.
Keywords :
"Speech","Sentiment analysis","Databases","Time-frequency analysis","Noise","Organizations","Speech recognition"
Conference_Titel :
Communication Systems and Network Technologies (CSNT), 2015 Fifth International Conference on
DOI :
10.1109/CSNT.2015.97