The Arabic language alphabets are important to every Muslims and their correct pronunciation is the key to recite the Quran correctly. The problem for non-native Muslims is that they do not know the Arabic language and hence cannot read alphabets correctly and especially their pronunciation remains very weak. There are many educational programs aim to teach the Arabic alphabets but there is an important missing part which is how to pronounce the alphabets and the application should give the user feedback and status of learning.
Our application aims to provide many features and focus especially on the way of pronunciation of the alphabets of the Quran. We provide this function to help all Muslims to read the Quran properly and with the correct pronunciation of the letters in an easy way by using deep learning with two-way interaction from the application to the user and vice versa.
How we built it
- Python language: In computer world there are many programming languages use it in many falids. Python is one of an important programming language that can be used on operating system. It can be used for processing many things such as text, images, numbers and audio. One failed of using python language is using it in goggle engine .Python contains multiple packet and library used for audio processing. This package will be very helpful in Speech Recognition. There are many libraries in Python used in audio processing.
2.Audio Classification Audio proccing has many problem one of the most problem faced it is Audio classification. The goal of audio classification is to defend the class that the audio is belong to. Before using the audio there are number of steps which are preprocessing the audio to extract and get the feature then using audio classification to get the audio and its feature in phase called then pass it by using a neural network to get the and identefie appropriate class.
3.Audio Segmentation Audio segmentation means make the audio as parts depend on set of propereties and characters. It is an important step by audio segmrntation we can divide the lengh of the audio to be short audio.
4.Android Studio Android Studio it’s a useful application which help the programmer to write coding , do editing on in it and many others services.
Filmora9 It is an application provide a service to the user which creates video, edit video, etc. we use this app to do the preprocessing for our audios which are cleaning the audios, separate the letter in the audios and remove noising in the background for each audio.
Jupiter notebook The Jupiter Notebook is web that allow the user to perform many things such as : creating , sharing visualizations. We use it to write and Run and share our codes written in python languages.
Challenges we ran into
Our challenge is to develop an appropriate audio model for the Arabic alphabet so that the model can act as a reference for the learners. The application aims to teach the user Arabic alphabet pronunciation in a correct way to read the Quran properly. The application offers the user the ability to record the voice so that the application can give feedback to the voice to the user to make sure that the pronunciation is correct for the alphabet. This application will solve many difficulties of non-Arabic Muslim's experience of learning.
Accomplishments that we're proud of
Providing assistance and benefit to all Muslims so that they can recite the Holy Quran to the fullest.
What we learned
We learned to implement the work with thoughtful steps and implement deep learning on the program
What's next for Deep Learning Based Holy Quran Recitation Assistant
We will define and measure the similarity between the user pronunciation of the Quran alphabet and the correct pronunciation and provide feedback to the user.