Speech Based Query Searching Technique and It’s Application In Library Management System

Abstract

Customary library inventories have turned into wasteful and badly designed in assisting library clients. Readers are spending too much time in looking through materials related to library by means of inventory copies. Pursuers need a smart and creative answer to sort out this issue using speech itself. Speech is the most effortless approach to speak with one another. Recently, Speech processing is mostly used in applications like devices used for security purposes, household appliances, cellular/mobile devices, Automatic Teller Machines (ATMs) ,computers and etc. The man machine interface is also created to associate people who are experiencing some kind of disabilities. Speech processing is nothing but the strategy to process and break down the speech signals. It has great advantages for disabled people and discovers their applications in our everyday lives. In this paper, we have connected different data mining models to distinguish the required course book which is available with using speech signals and furthermore findings in a book which is frequently issued to readers. It is very much useful to deaf peoples in society. In this we applied Mel-frequency cepstral Coefficients (MFCC) to extract the features from speech signal of isolated spoken words and applied GMM and DNN models for discovering whether required book is present or not and observed is DNN gives the great accuracy and We are also converting the given speech into text format and apply association rules to find the book that is frequently issued.