A STUDY OF ARTIFICIAL INTELLIGENCE AS A REAL TIME HOST

Abstract

This paper deals with application of the electronic-dj which brings out a friendly relationship between the listener and the tuner. Off course, these both sequences of actions are done by a single individual. In this, the selection of the song is been accessed through speech recognition. The main purpose of implementing this model is, to favor and provide entertainment to the user. This paper presents a Novel approach to speech recognition. Currently, most speech recognition system is based on Hidden Markov models (HMMs), a statistical framework that supports both acoustic and temporal modeling. HMMs make a number of suboptimal modeling assumptions that limit their potential effectiveness.  Neural Networks avoid many of these assumptions, while they can learn complex functions and generalize effectively. Thus this method is tested over a standard speech data base and the results are presented. In this paper we describe a speaker independent speech recognition system. The module performs recognition using microphone. This model establishes the environment where the user can interact with the system for his favorite song selection from the songs listed in the database by his oral communication. Speaker independent speech recognition is important for successful development of speech recognizers in most real world applications like an E-DJ. While speaker dependent speech recognizers have achieved close to 100% accuracy, the speaker independent speech recognition systems have poor accuracy not exceeding 75%. This model gave 85% of true results.

Index Terms: DJ, Neural Network, Music retrievals, Speech recognition, Intelligent Systems and Approach.