Browsing by Author "Muhammed Aliyu Suleiman"

Now showing 1 - 4 of 4

Development of Hausa dataset a baseline for speech recognition
(Data in Brief, 2022-01-10) Umar Adam Ibrahim; Moussa Mahamat Boukar; Muhammed Aliyu Suleiman
The Hausa language read-speech dataset was created by recording native Hausa speakers. The recording took place at Nile university of Nigeria audio studio and radio broadcasting studio. The recorded dataset was segmented into unigram and bigram. The Hausa speech dataset contain 47hr of recorded audio speech. The dataset can be used for automatic speech recognition, speech synthesis, Text-to-Speech and speech-to-text application
Graphic User Interface for Hausa Text-to-Speech System
(IEEE, 2022-02-02) Umar Adam Ibrahim; Moussa Mahamat Boukar; Muhammed Aliyu Suleiman
Natural language processing and Digital signal processing are broadly used methods used to enable systems to understand commands and manipulate speech or text. Most of the Text-to-speech done was for major languages such as English, French and others, with no or little for African languages like Hausa, which are termed under resource languages. In this paper, we developed a graphical user interface for the Hausa Text-to-Speech system. This system converts Hausa text to Hausa audio sound, by processing and analyzing it using natural language processing and Digital Signal Processing. Our graphical user interface, aid in converting entered Hausa language text into Hausa speech.
Reinforcement Learning for Testing Android Applications
(International Conference on Multidisciplinary Engineering and Applied Science, 2023-02-02) Moussa Mahamat Boukar; Muhammed Aliyu Suleiman; Ibrahim Anka Salihu
This paper offers a review of current research studies that use reinforcement learning (RL) to test Android applications. The primary purpose of this study is to simplify future research by collecting and investigating the current state of Android app testing approaches using the RL technique. We provide a well-defined criterion comprising of seven key points. The key points are: addressed problems, reasons for using the RL technique, RL algorithms, supported events, testing techniques, validation, and evaluation methods. In the literature, we have analyzed various techniques to evaluate their efficiency. This study showed that model-based testing is the most commonly used testing technique. Q-learning is the best algorithm in terms of predictive accuracy. We identified that code coverage is the most widely used evaluation metric and comparison with other tools and techniques is the preferred validation approach.
Test Case Generation Approach for Android Applications using Reinforcement Learning
(Engineering, Technology & Applied Science Research, 2024-04-27) Moussa Mahamat Boukar; Muhammed Aliyu Suleiman; Ibrahim Anka Salihu
Mobile applications can recognize their computational setting and adjust and respond to actions in the context. This is known as context-aware computing. Testing context-aware applications is difficult due to their dynamic nature, as the context is constantly changing. Most mobile testing tools and approaches focus only on GUI events, adding to the deficient coverage of applications throughout testing. Generating test cases for various context events in Android applications can be achieved using reinforcement learning algorithms. This study proposes an approach for generating Android application test cases based on Expected State-Action-Reward-State-Action (E-SARSA), considering GUI and context events for effective testing. The proposed method was experimentally evaluated on eight Android applications, showing 48- 96% line of code coverage across them, which was higher than Q-testing and SARSA.