Word based emotion conversion in Hindi language

Abstract

The main system to communicate with computers is the use of natural language in text form. To add naturalness and intelligibility, the speech form is becoming an important method to communicate with computers and other machines.Human-machine and human-robot dialogues in the next generation will be conquered by natural speech, which is fully impulsive and thus obsessed by emotion. Emotion adds expressiveness to the natural language speech. There is a great zeal to research in this field. In this paper we have proposed an algorithm for word based emotion conversion of neutral speech into emotional speech like 'happy' and 'sad' for Hindi language. This emotion conversion algorithm is based on the segmentation of the spoken utterance into words and the pitch differences of these words between different emotions. The segmentation of words in the spoken utterance is done using another algorithm of word boundary detection, which gives details of start of the word, end of the word and the no. of words in a sentence. This algorithm of word boundary detection is devised for Hindi language and is based on two main prosodic features 'pitch' and 'intensity'.

A Machine Learning Approach for Emotion Classification in Bengali Speech

“…Emotional understanding promotes mutual understanding and contributes to a fulfilling social life [1]. In today's technologically advanced world, where voice commands and manual instructions are exchanged between humans and machines, the ability of artificial intelligence or machine learning to identify and assess someone's emotional state from their voice has emerged as a pressing challenge [2] [3]. There is little study on emotion classification in non-English languages like Bengali, which presents particular difficulties for speech analysis.…”