Emotion detection from male speech in computer games

This paper presents the experimental results of our method of classifying emotional states of neutral and anger of male voice from artificial pidgin utterances using Support vector machine (SVM). The objective of the paper is to demonstrate that a new genre of languages called Game pidgin language (GPL) that can not only be used to capture real time speech recognition, but can also generate response from the non-player-character. The emotion which is a floating point value can be used to generate the corresponding emotive response from the non-player- character (NPC).