1.1 Introduction
Speech Recognition software can "read" text from a document, Web page or e-Book, and also listen to voices of the system user, thereby generating synthesized speech through a computer's speakers. Genetic algorithm based on natural genetics; therefore they share the same names. The genetic algorithms is a stochastic search technique (stochastic search use probability to help guide their search) inspired by the mechanics of natural selection and natural genetics (Goldberg et al., 1989). The basic idea behind the genetic algorithms is to maintain a population of strings or chromosome, which are encoding of a potential solution to the problem being investigated. Each chromosome is tested using a fitness function to know the good solution of the problem. The strings of artificial genetic system are analogous to chromosome in biological system. The chromosomes are composed of features, or detectors that are called genes. This may take on some number of values, called alleles.
As a prelude to other parts of this study, this chapter will discuss the background upon which this study was initiated, the statement of problems that led to this study, the Aim and Objectives of the study. Others are Significance of the study, Scope of work, Limitations of the Study and Definition of technical terms.
1.2 Background of Study
In recent years, study on Genetic Algorithm can be found in many research papers (Chu, 2003a; Chen, 2003; Chu, 2003b). They demonstrated different characteristics in Genetic Algorithm than others. For example, parallel search based on random multi-points, instead of a single point, was adopted to avoid being limited to local optimum. In the operation of Genetic Algorithm, it only needs to establish the objective function without auxiliary operations, such as differential operation. Therefore, it can be used for the objective functions for all types of problems.
Speech signals are composed of a sequence of sound. These sound and the transitions between them serve as symbolic representation of information. The arrangement of these sounds (symbols) is governed by the rule of language. The study of these rules and their implications in human communication is the domain of linguistic. The study and classification of the sounds of speech is called phonetics. Speech can be represented in term of its message content or information. An alternative way of characterizing speech is in terms of the signal carrying the message information, i.e., the acoustic waveform (Holmes et al., 2001). Speech is one of the most important tools for communication between human and his environment, therefore manufacturing of Automatic System Recognition (ASR) is desire for him all the time (Rabiner et al., 1978).
Artificial neural network has better speech recognition speed and less calculation load than others, it is suitable for chips with lower computing capability. Therefore, artificial neural network was adopted in this study as speech recognition platform. Most artificial neural networks for speech recognition are back-propagation neural networks. The local optimum problem (Yeh, 1993) with Steepest Descent Method makes it fail to reach the highest recognition rate. In this study, Genetic Algorithm was used to improve the drawback.
Consequently, the mission of this chapter is the experiment of speech recognition under the recognition structure of Artificial Neural Network (ANN) which is trained by the Genetic Algorithm (GA). This chapter adopted Artificial Neural Network (ANN) to recognize Mandarin digit speech. Genetic algorithm (GA) was used to complement Steepest Descent Method (SDM) and make a global search of optimal weight in neural network. Thus, the performance of speech recognition was improved. The nonspecific speaker speech recognition was the target of this chapter. The experiment in this chapter would show that the GA can achieve near the global optimum search and a higher recognition rate would be obtained. Moreover, two method of the computation of the characteristic value were compared for the speech recognition.
However, the drawback of GA used to train the ANN is that it will waste many training time. This is becasue that the numbers of input layer and output layer is very large when the ANN is used in recognizing speech. Hence, the parameters in the ANN is emormously increasing. Consequently, the training rate of the ANN becomes very slow. It is then necessary that other improved methods must be investigated in the future research.
In natural systems, one or more chromosome combined to form the total genetic prescription for the construction and operation of some organism. The total genetic package (structure) is called the genotype. The organism formed by the interaction of the total genetic package with its environment is called the phenotype (Mitchell, 1996). Genetic algorithm based on natural genetics; therefore they share the same names. The genetic algorithms is a stochastic search technique (stochastic search use probability to help guide their search) inspired by the mechanics of natural selection and natural genetics (Goldberg et al., 1989). The basic idea behind the genetic algorithms is to maintain a population of strings or chromosome, which are encoding of a potential solution to the problem being investigated. Each chromosome is tested using a fitness function to know the good solution of the problem. The new population is created by selecting chromosome from the old population. The new population is re-evaluated and the processes continue until the solution is found (Mitchell et al., 1996). The strings of artificial genetic system are analogous to chromosome in biological system. The chromosomes are composed of features, or detectors that are called genes. This may take on some number of values, called alleles.
1.3 Statement of the Problem
The importance of texts cannot be overemphasized. Hardly can anyone pass a message without including one form of text or the other. This is a problem for the visually impaired. They find it hard to read through the texts especially when the font-size is small. This has led to the development of a Speech Recognition conversion system. For those with learning disabilities, some in literary levels, they often get frustrated trying to browse the internet because so much of it is in text form.
Also in some already developed speech synthesizers, the problem area in speech synthesis is very wide. There are several problems in speech pre-processing, such as numerals, abbreviations, and acronyms. This system will help solve the problems by using well written synthesis algorithm for the conversion.
1.4 Aim and Objectives of Study
The aim of the study is to Design and Implement a Speech Recognition System using Genetic Algorithm. In achieving this aim, the following specific objectives were laid out as follows:
- To understand the speech recognition and its fundamentals.
- To design and implement a Speech synthesizer that will recognize voice sound.
- To design and implement a System that can listen to speech in any frequency that user specifies.
- To develop an application software that can mainly be used for: Speech Recognition, Speech Generation, Text Editing and Tool for operating Machine through voice.
1.5 Significance of Study
The following are the relevance of this study, which are stated as follows:
- The application will build a platform to aid people with disabilities especially on reading and also help get information easily without any stress.
- The study will serve as a foundation and guide to other research students interested in researching on Speech Recognition systems.
1.6 Scope of the Study
The study focuses on the Design and Implementation of Speech Recognition System using Genetic Algorithm.
1.7 Limitations of the Study
During the course of this study, many things militated against its completion, some of which are:
- Time Constraint: The time frame given to accomplish this project was very short due to school academic calendar and it was carried out under pressure which made the researcher not to implement some necessary features.
- Research material: availability of research material is a major setback to the scope of the study.
- Frequent power failure: This made the researcher append more money on fuel to ensure sustainable power.
- Financial Constraint: Insufficient fund tends to impede the efficiency of the researcher in sourcing for the relevant materials, literature or information and in the process of data collection (internet).
1.8 Definition of Terms
Learning Objects: In the context of this research, a learning-object-is-a-resource with a clear educational-application. It is in digital form − examples: a Microsoft Word Document or a PDF document.
Catalog: This research applies the word ‘catalog’ as a container that presents all the SICT projects logged/stored in an organized tabular format, showing relevant information about each project. It can be compared to a library catalog containing all the information about books − their titles, author, number of pages, published year, ISBN etc.
Repository: A repository (in this case; SICT repository) is an archive, storehouse or container that allows storage, cataloging, accessing all SICT undergraduate projects and thesis, and viewing each object’s content.
Identifier: An identifier is simply a unique identity attached to one object, it aids referencing such object directly when its identifier is requested. An identifier can be a serial number, identity number (also written as ‘ID’) or an index number.
Object: An object in the context of this research refers to a project record returned by a database − queried using an Object Oriented Programming (OOP) approach. An object (i.e. a project record) returns or contains the project topic, student name, abstract, year of project submission and the date and time the object record was created.
Database: A database is a repository that allows storage, retrieval and manipulation of data. A database can be used efficiently with the aid of a Database Management System (DBMS) − a set of tools that allow storage, access, retrieval and maintenance of data stored in a database, examples of DBMS include XAMPP, WAMP, LAMP and AppServ to mention a few.
Web-Browser: A web-browser, internet browser or browser is software program that interprets the codes written in markup languages in graphic and visual (like images, text, audio or animation) form. This allows users to easily request and access a website or to search for information through a search engine (either Google, Ask, Bing and AOL among others).
Keyword: A phrase or just one word that is used to search for a certain result or set of results.
User: A user is system, application, request or person that can use a computer or software to perform a specific task.
Metadata: A set of data that describes and gives information about other data.
Dataset: A collection of related game plans of information that is made out of disengaged segments however can be controlled as a unit by a PC.
Relational Database: A database composed to see association among relations of information delineates tables from which data can be gotten to or reassembled in different courses without rearranging the database tables.
