MyMedR

Displaying all 8 publications

Abstract:

Sort:

CNN-based noise reduction for multi-channel speech enhancement system with discrete wavelet transform (DWT) preprocessing

Cherukuru P, Mustafa MB

PeerJ Comput Sci, 2024;10:e1901.
PMID: 38435554 DOI: 10.7717/peerj-cs.1901

Speech enhancement algorithms are applied in multiple levels of enhancement to improve the quality of speech signals under noisy environments known as multi-channel speech enhancement (MCSE) systems. Numerous existing algorithms are used to filter noise in speech enhancement systems, which are typically employed as a pre-processor to reduce noise and improve speech quality. They may, however, be limited in performing well under low signal-to-noise ratio (SNR) situations. The speech devices are exposed to all kinds of environmental noises which may go up to a high-level frequency of noises. The objective of this research is to conduct a noise reduction experiment for a multi-channel speech enhancement (MCSE) system in stationary and non-stationary environmental noisy situations with varying speech signal SNR levels. The experiments examined the performance of the existing and the proposed MCSE systems for environmental noises in filtering low to high SNRs environmental noises (-10 dB to 20 dB). The experiments were conducted using the AURORA and LibriSpeech datasets, which consist of different types of environmental noises. The existing MCSE (BAV-MCSE) makes use of beamforming, adaptive noise reduction and voice activity detection algorithms (BAV) to filter the noises from speech signals. The proposed MCSE (DWT-CNN-MCSE) system was developed based on discrete wavelet transform (DWT) preprocessing and convolution neural network (CNN) for denoising the input noisy speech signals to improve the performance accuracy. The performance of the existing BAV-MCSE and the proposed DWT-CNN-MCSE were measured using spectrogram analysis and word recognition rate (WRR). It was identified that the existing BAV-MCSE reported the highest WRR at 93.77% for a high SNR (at 20 dB) and 5.64% on average for a low SNR (at -10 dB) for different noises. The proposed DWT-CNN-MCSE system has proven to perform well at a low SNR with WRR of 70.55% and the highest improvement (64.91% WRR) at -10 dB SNR.
Emotional speech acoustic model for Malay: iterative versus isolated unit training

Mustafa MB, Ainon RN

J Acoust Soc Am, 2013 Oct;134(4):3057-66.
PMID: 24116440 DOI: 10.1121/1.4818741

The ability of speech synthesis system to synthesize emotional speech enhances the user's experience when using this kind of system and its related applications. However, the development of an emotional speech synthesis system is a daunting task in view of the complexity of human emotional speech. The more recent state-of-the-art speech synthesis systems, such as the one based on hidden Markov models, can synthesize emotional speech with acceptable naturalness with the use of a good emotional speech acoustic model. However, building an emotional speech acoustic model requires adequate resources including segment-phonetic labels of emotional speech, which is a problem for many under-resourced languages, including Malay. This research shows how it is possible to build an emotional speech acoustic model for Malay with minimal resources. To achieve this objective, two forms of initialization methods were considered: iterative training using the deterministic annealing expectation maximization algorithm and the isolated unit training. The seed model for the automatic segmentation is a neutral speech acoustic model, which was transformed to target emotion using two transformation techniques: model adaptation and context-dependent boundary refinement. Two forms of evaluation have been performed: an objective evaluation measuring the prosody error and a listening evaluation to measure the naturalness of the synthesized emotional speech.
Fulltext Mobile learning: research context, methodologies and future works towards middle-aged adults - a systematic literature review

Mohtar S, Jomhari N, Mustafa MB, Yusoff ZM

Multimed Tools Appl, 2023;82(7):11117-11143.
PMID: 36035325 DOI: 10.1007/s11042-022-13698-y

Over the past several years, mobile learning concepts have changed the way people perceived on mobile devices and technology in the learning environment. In earlier days, mobile devices were used mainly for communication purposes. Later, with many new advanced features of mobile devices, they have opened the opportunity for individuals to use them as mediated technology in learning. The traditional way of teaching and learning has shifted into a new learning dimension, where an individual can execute learning and teaching everywhere and anytime. Mobile learning has encouraged lifelong learning, in which everyone can have the opportunity to use mobile learning applications to gain knowledge. However, many of the previous studies on mobile learning have focused on the young and older adults, and less intention on middle-aged adults. In this research, it is targeted for the middle-aged adults which are described as those who are between the ages of 40 to 60. Middle-aged adults typically lead very active lives while at the same time are also very engaged in self-development programs aimed at enhancing their spiritual, emotional, and physical well-being. In this paper, we investigate the methodology used by researchers based on the research context namely, acceptance, adoption, effectiveness, impact, intention of use, readiness, and usability of mobile learning. The research context was coded to the identified methodologies found in the literature. This will help one to understand how mobile learning can be effectively implemented for middle-aged adults in future work. A systematic review was performed using EBSCO Discovery Service, Science Direct, Google Scholar, Scopus, IEEE and ACM databases to identify articles related to mobile learning adoption. A total of 65 journal articles were selected from the years 2016 to 2021 based on Kitchenham systematic review methodology. The result shows there is a need to strengthen research in the field of mobile learning with middle-aged adults.
A speech pronunciation practice system for speech-impaired children: A study to measure its success

Salim SS, Mustafa MB, Asemi A, Ahmad A, Mohamed N, Ghazali KB

Res Dev Disabil, 2016 Sep;56:41-59.
PMID: 27262125 DOI: 10.1016/j.ridd.2016.05.013

BACKGROUND: The speech pronunciation practice (SPP) system enables children with speech impairments to practise and improve their speech pronunciation. However, little is known about the surrogate measures of the SPP system.
AIMS: This research aims to measure the success and effectiveness of the SPP system using three surrogate measures: usage (frequency of use), performance (recognition accuracy) and satisfaction (children's subjective reactions), and how these measures are aligned with the success of the SPP system, as well as to each other.
METHODS AND PROCEDURES: We have measured the absolute change in the word error rate (WER) between the pre- and post-training, using the ANOVA test. Correlation co-efficiency (CC) analysis was conducted to test the relation between the surrogate measures, while a Structural Equation Model (SEM) was used to investigate the causal relations between the measures.
OUTCOMES AND RESULTS: The CC test results indicate a positive correlation between the surrogate measures. The SEM supports all the proposed gtheses. The ANOVA results indicate that SPP is effective in reducing the WER of impaired speech.
CONCLUSIONS AND IMPLICATIONS: The SPP system is an effective assistive tool, especially for high levels of severity. We found that performance is a mediator of the relation between "usage" and "satisfaction".
Fulltext Severity-based adaptation with limited data for ASR to aid dysarthric speakers

Mustafa MB, Salim SS, Mohamed N, Al-Qatab B, Siong CE

PLoS One, 2014;9(1):e86285.
PMID: 24466004 DOI: 10.1371/journal.pone.0086285

Automatic speech recognition (ASR) is currently used in many assistive technologies, such as helping individuals with speech impairment in their communication ability. One challenge in ASR for speech-impaired individuals is the difficulty in obtaining a good speech database of impaired speakers for building an effective speech acoustic model. Because there are very few existing databases of impaired speech, which are also limited in size, the obvious solution to build a speech acoustic model of impaired speech is by employing adaptation techniques. However, issues that have not been addressed in existing studies in the area of adaptation for speech impairment are as follows: (1) identifying the most effective adaptation technique for impaired speech; and (2) the use of suitable source models to build an effective impaired-speech acoustic model. This research investigates the above-mentioned two issues on dysarthria, a type of speech impairment affecting millions of people. We applied both unimpaired and impaired speech as the source model with well-known adaptation techniques like the maximum likelihood linear regression (MLLR) and the constrained-MLLR(C-MLLR). The recognition accuracy of each impaired speech acoustic model is measured in terms of word error rate (WER), with further assessments, including phoneme insertion, substitution and deletion rates. Unimpaired speech when combined with limited high-quality speech-impaired data improves performance of ASR systems in recognising severely impaired dysarthric speech. The C-MLLR adaptation technique was also found to be better than MLLR in recognising mildly and moderately impaired speech based on the statistical analysis of the WER. It was found that phoneme substitution was the biggest contributing factor in WER in dysarthric speech for all levels of severity. The results show that the speech acoustic models derived from suitable adaptation techniques improve the performance of ASR systems in recognising impaired speech with limited adaptation data.
Klebsiella sp. FIRD 2, a TBT-resistant bacterium isolated from contaminated surface sediment along Strait of Johor Malaysia

Abubakar A, Mustafa MB, Johari WLW, Zulkifli SZ, Ismail A, Mohamat-Yusuff FB

Mar Pollut Bull, 2015 Dec 15;101(1):280-283.
PMID: 26434791 DOI: 10.1016/j.marpolbul.2015.09.041

A possible tributyltin (TBT)-degrading bacterium isolated from contaminated surface sediment was successfully identified as Klebsiella sp. FIRD 2. It was found to be the best isolate capable of resisting TBT at a concentration of 1000 μg L(-1). This was a concentration above the reported contaminated level at the sampling station, 790 μg L(-1). Further studies revealed that the isolate was Gram negative and resisted TBT concentrations of up to 1500 μg L(-1) in a Minimal Salt Broth without the addition of any carbon source within the first 48 h of incubation. It is expected that additional work could be conducted to check the degradation activity of this new isolate and possibly improve the degradation capacity in order to contribute to finding a safe and sustainable remediation solution of TBT contamination.
ICU nurses' safety performance related to respect for safety and management commitment: A cross-sectional study

Al-Bsheish M, Jarrar M, Mustafa MB, Zubaidi F, Ismail MAB, Meri A, et al.

Contemp Nurse, 2022;58(5-6):446-459.
PMID: 35856481 DOI: 10.1080/10376178.2022.2104740

BACKGROUND: Healthcare work is one of the most accident-prone occupations globally. Nurses, especially those who work in Intensive Care Units (ICU), are very likely to experience mishaps on the job due to the complicated duties they perform. Safety performance through compliance and participation in safety is a proactive approach and a critical tool to measure the protection of employees, like these, in the workplace. Although interest in this tool has increased among hospital administrators and managers, scientific research has been limited in this area.
AIMS: The study's purposes were twofold: (1) to explore the effect of perceived respect safety on the safety performance of ICU nurses and (2) to explore the mediation effect of Management Commitment to Safety (MCS) between the relationship of perceived respect safety and safety performance.
METHODS: Eight public hospitals from the Jordanian Ministry of Health (JMoH) were selected randomly using cluster sampling, and their ICU nurses were surveyed. A total of 285 nurses completed questionnaires. The SmartPLS3 bootstrapping technique was used to analyse data.
RESULTS: The results established that the perceived respect for the safety of nurses has a significant and positive effect on their safety compliance (β = .39, p
Intra-rater and inter-rater reliability of total faulty breathing scale using visual observation and videogrammetry methods

Mohan V, Perri M, Paungmali A, Sitilertpisan P, Joseph LH, Jathin R, et al.

J Bodyw Mov Ther, 2017 Jul;21(3):694-698.
PMID: 28750986 DOI: 10.1016/j.jbmt.2016.10.007

Faulty breathing is an aspect of alteration in the normal fundamental pattern of breathing. The available existence of scales in assessing faulty breathing has not frequently been used. Measurement errors in assessing and quantifying breathing patterns may originate from unclear directions and variation between observers. This study determined the measure reliability of the Total Faulty Breathing Scale (TFBS) for quantifying breathing patterns. Twenty seven participants were recruited comprising healthy and unhealthy subjects. Two examiners assessed their breathing patterns using the TFBS on two different occasions with visual observation and a videogrammetry method. Evaluation of the observational breathing pattern method for intra-rater and inter-rater showed agreement of 96.30% and a kappa score of greater than 0.78, which indicated substantial agreements. The videogrammetry method showed a percent agreement of (100%) with a kappa score of (1.00). This study indicates that the TFBS is a considerably reliable tool for evaluating breathing patterns with both visual observation and a videogrammetry method.