Speech Recognition Focusing on the Holy Qur’an (AL-FURQAN APPLICATION)

  • : Ms Word Format
  • : 70 Pages
  • : ₦3000
  • : 1-5 Chapters
  •  
  • Click to DOWNLOAD Materials

Speech Recognition Focusing on the Holy Qur’an (AL-FURQAN APPLICATION)

ABSTRACT

This project report describes a mobile android based Qur’an recognition application. This application consists of features that can help Muslims across the globe to stay in touch with their religious obligations and also to improve their knowledge of the holy Qur’an. The application comes with features like recognition of Qur’an verses, English translation of recognised verse(s) and a compass showing the direction to pray towards. Information and data gathering during the development of the system was done through interviews and review of the appropriate literature. This dissertation also consists of the activity diagrams, the application architecture, uses cases and entity relationship diagrams. The technologies used in the execution of the objectives of this research are android studio as the IDE of choice and SQLITE for the database. Upon compelling appraisal and system verification, the writer sees that the application can be used modestly, productively, effectively and efficiently.

CHAPTER 1: INTRODUCTION

1.1     Overview

Speech recognition has become a very dominant discipline in computer science. Speech recognition involves the utterance of words by an individual which then gets translated to machine recognizable words or actions (Sharma U. and Mishra A., 2015). Speech recognition gained dominance as a result of current advancements in research for fields like artificial intelligence and machine learning, this advancement has led to the many commercialized versions of speech recognition like the Alexa by Amazon, Siri by Apple, Cortana by Microsoft, Talk to Google by Alphabet Inc. etc. These commercialized versions of speech recognition have gained dominance in the mobile sector of the industry with some in devices like the laptop, portable speakers and wearables etc. Mobile devices are a must have in current times as they are the main source of aiding in effective communication between humans, it is safe to say that mobile devices are the key tools in connecting people across the globe and this in turn has led to the creation of opportunities for business and personal purposes. The business demand of mobile devices that come with speech recognition systems is on the rise and this is so because of the application of this technology to several languages, they are currently vast amounts of languages for which speech recognition models have been built and these models are used to aid different aspects of life for the respective languages, these include general dictation, text to speech transcription and language translation. language translation using speech recognition involves recognizing speech in a particular language and then converting that speech input to its equivalent meaning in another language. Speech recognition consists of a combination of computing, linguistics and mathematics, the first approach towards achieving speech recognition is digital sampling of speech input which means the device will take an analogue signal from a human being and then convert it to a digital signal, the next approach is the acoustic signal processing which is the conversion of the captured signal into words, the final approach is making the computer understand these words which can be achieved by applying different processes like the Hidden Markov models, Artificial Neural Networks and Dynamic Time Wrapping. There are several methods of performing speech recognition and here below are some of the different types and levels of speech recognition.

  • Pattern and feature analysis: The machine break the words down into bits and then understands them by the vowels they contain.
  • Pattern matching: The machine understands base on what it was taught without any conscious analysis.
  • Language modelling and statistical analysis: The machine uses the understanding of grammar, words and sounds. It makes recognition fast and it enhances accuracy.
  • Artificial neural networks: It is the most reliable form of speech recognition presently; it understands words and sounds after it undergoes extensive training. The system is designed to work independently on devices e.g. Siri by Apple and the Alexa by Amazon.

This chapter contains the background and motivation, the statement of problem, aims and objectives, the significance of undertaking this project, risk assessment and the scope/project organization.

1.2     Background and Motivation

Speech recognition gained a lot more traction in late 2010, with the introduction of Siri by Apple Inc. Speech recognition at its early stages during the 20th century was only able to recognize digits spoken by a single voice instead of human speech (Medium, 2018). Few years later a speech recognition system was created and it was able to understand sixteen (16) various voices (Medium, 2018). The major purpose of speech recognition is the establishment of an effective means for communication between humans and machines and this has helped majorly in breaking down barriers of hardship by making communication between man and machine easier with every new technological breakthrough in the field. The Holy Quran is amongst the greatest religious testaments dear to a Muslim and of which about two (2) billion of them exist, and the Holy Quran is written in a different variation of Arabic language apart from the normal variation known to Man and this variation of Arabic language has no effective speech recognition system readily available for use. About two third (2/3) of the Muslim population around the globe have a similar problem which is understanding what the verses stand for in a suitable language known to them and also, they deal with forgetfulness of certain aspects of the Holy Quran due to its enormous coverage and depth. With this application, a great majority of the Muslim Population will have a predisposed speech recognition system that it can always refer to in order to solve the aforementioned issues and further boost their knowledge of the Holy Quran.

 

1.3     Statement of the Problem

The Holy Quran is one of the most precious books to Muslims across Nigeria and the globe as a whole, but it is a known fact that most Muslims only delve into the Arabic aspect of the Holy Quran without regard for the English aspect or at least the meaning of what they learn about the Holy Quran. This creates a gap between the understanding of the Holy Quran and the learning of the Arabic aspect. The world boasts of about three (3) billion Muslims worldwide and this problem exists in about one (1) in one-hundred (100) Muslims across the globe which makes this a very serious concern that needs a technological solution. This problem at hand is currently solved by employing a private tutor to educate the concerned parties but with this application a technological solution is provided to this problem which brings about efficiency in many ways i.e. financial, time and so on.

1.4     Aim and Objectives

The system is a speech recognition tool where the user inputs verses of the Qur’an in form of speech and the system should be able to recognize the verse and it should be able to display the verse, the transliteration and its translation from the Qur’an, it will also contains a Kiblah locator (prayer direction compass) and to develop this system the following objectives are the main priority;

  • Collection of data and resources.
  • Extraction of features.
  • Use of speech classifiers.
  • Application implementation and development.

The main aim of this project is to aid Muslims across the globe in their religious endeavours, presently only applications that recognize Arabic language are available, and although the Qur’an is written in Arabic it is classified as a totally different dialect than the normal Arabic dialect that exists. Tutors for the Holy Scripture are available but they cost money and humans in their true nature can be unreliable but with this application installed on your android device you can take it anywhere and use in the hopes of gaining new insights and knowledge of the Holy Qur’an and to aid in locating where to face in order to perform the obligatory prayers in any location unknown to the user of this application.

1.5     Significance of the Project

The implementation of this project has the potential to benefit the academia, the Nigerian Muslim society, philanthropy, and the Muslim society across the globe. This research adds more knowledge to the Nigerian Muslim society by providing hitherto information on the

Holy Qur’an to its members in a vast amount. Also, a concise report on each phase of the implementation of the project and room for further development will be available. This research project stands as the first of its kind to be written, and thus opens a fresh avenue for research in this area. Al-Furqan android application contributes even more to the Muslim Society across the globe. With Al-Furqan, the Muslim society plays a better part in understanding its religion diligently. Al-Furqan provides a better medium for aiding less knowledgeable Muslims in understanding their religion. This creates a better and a more knowledgeable Muslim society across the globe. This positive development could extend to other countries and spread across all its axes, thereby making the world a better place to live in. This project would aid in increasing the number of Qur’an literate Muslims globally. Speech recognition applications are getting mainstreamed into the human culture, it is becoming a regular day to day activity to an average mobile phone user. Al-Furqan represents a new era in religious aspect of computing and technology.

1.6     Project Risks Assessment

Table 1: Project Risk Assessment

Risk(s) Possible solution(s)
Inability to carry out research due to loss of hardware/software resources. Be aware of and observe school security procedures.
Loss     of         work    due      to             equipment failure/loss. Weekly data backup.
Software availability (Unavailability of API’s). Alternative API’s will be checked for.
Unwillingness             to         pass             sensitive information/business model. Non-disclosure agreement signing.
Inadequate supervision. Report to a higher authority like the HOD.
Health related issues. Report to the project management team.

1.7     Scope/Project Organization

The work done in this project is organized into different chapters; there is a total of five chapters in this dissertation. Chapter 1 includes the background information on the speech recognition field, the statement of the problem at hand, the problem objectives in addition with the aims of the project, lastly, we have the project risk assessment and the scope of the project. Chapter 2 provides the historical overview of the field of choice and also related works done in the field chosen. Chapter 3 consists of the methods/approaches and the overall framework taken in building the application, the requirements of the application both functional and non-functional, hardware and software requirements are also included in this chapter, the activity diagrams, use case diagrams, ERDs and the user interface design. Chapter 4 covers the system implementation, testing on various levels and their various results, the main features of the application will be mentioned and discussed, problems encountered during implementation and how they were solved, error reports and solutions will be provided also. Chapter 5 covers the overview of the whole dissertation, the objective assessment, suggested improvements, limitations and challenges encountered during implementation, recommendations, and summary. Other areas covered at the end are references, and appendices. The Gantt chart and work plan depicting the project schedule can be found in the appendices section.

Speech Recognition Focusing on the Holy Qur’an (AL-FURQAN APPLICATION)

Sharing is caring!

Leave a Reply