Phonetic dictionaries are essential components of large-vocabulary speaker-independent speech recognition systems. This paper presents a rule-based technique to generate phonetic dictionaries for a large vocabulary Arabic speech recognition system. The system used conventional Arabic pronunciation rules, common pronunciation rules of Modern Standard Arabic, as well as some common dialectal cases. The paper gives in detail an explanation of these rules as well as their formal mathematical presentation. The rules were used to generate a dictionary for a 5.4 hour corpus of broadcast news. The rules and the phone set were tested and evaluated on an Arabic speech recognition system. The system was trained on 4.3 hours of the 5.4 hours of Arabic broadcast news corpus and tested on the remaining 1.1 hours. The phonetic dictionary contains 23,841 definitions corresponding to about 14232 words. The language model contains both bi-grams and tri-grams. The Word Error Rate (WER) came to 9.0%.
BackgroundThe Gulf Cooperation Council countries are witnessing unprecedented changes due to fast economic development and population growth. The aims of this study were twofold: first, to estimate the prevalence of diabetes and its comorbidities; second, to examine the association of sociodemographic risk factors and healthcare service utilisation with diabetes.MethodsData from the World Health Survey Plus (WHS+) from Kuwait, Oman, Saudi Arabia and the United Arab Emirates were used. The WHS+ is a nationally representative household survey of the adult population, conducted between 2008 and 2009. Both logistic regression and zero-inflated Poisson models were applied to examine the associations of risk factors, comorbidity and treatment with self-reported diabetes.ResultsThe highest level of diabetes was observed in Kuwait, with 40.8% among the oldest age group. High body mass index, older age and low education were all associated with diabetes in all settings. High levels of comorbidity existed within the diabetic population. Over 50% of diabetics in all countries reported having at least one chronic condition. In Kuwait and Saudi Arabia, one in five diabetics reported having two or more comorbidities. Treatment prevalence was above 80% across all sociodemographic categories.ConclusionThe burden of diabetes, although high, is not uniform across populations in the four Gulf countries. Differential exposure to risk, such as unhealthy lifestyles, may be creating a disadvantage for certain populations and influencing the co-occurrence of chronic conditions. In response, a multifaceted and patient-centred approach is needed at all levels of healthcare to control and prevent non-communicable diseases.
Phonetic dictionaries are essential components of large-vocabulary natural language speakerindependent speech recognition systems. This paper presents a rule-based technique to generate Arabic phonetic dictionaries for a large vocabulary speech recognition system. The system used classic Arabic pronunciation rules, common pronunciation rules of Modern Standard Arabic, as well as morphologically driven rules. The paper gives in detail an explanation of these rules as well as their formal mathematical presentation. The rules were used to generate a dictionary for a 5.4 hours corpus of broadcast news. The phonetic dictionary contains 23,841 definitions corresponding to about 14232 words. The generated dictionary was evaluated on an actual Arabic speech recognition system. The pronunciation rules and the phone set were validated by test cases. The Arabic speech recognition system achieves word error rate of %11.71 for fully diacritized transcription of about 1.1 hours of Arabic broadcast news.
This work presents a critical analysis for three models group of methanogen potential prediction. The first group allows determination of the methane productivity of substrates, through three models (BMPthCOD, BMPthAtC and BMPthOFC). The BMPthCOD is suitable for a first approximation calculation. BMPthAtC and BMPthOFC are more accurate; however, require a complex characterization of substrates. The second models group predicts the cumulative methane production using seven models. The analysis shows that the Artificial Neuron Network (ANN) is more accurate; moreover, it allows carrying out an optimization of the cumulative methane production. The third group of models is particularly involved in the determination of daily flow of methane by a biodigester. The Hashimoto model, which uses the operating parameters, has been identified as the most suitable.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.