检索结果-内蒙古大学图书馆

2022 International Conference on Information Science and Communications Technologies, ICISCT 2022

作者： Kurbonov, Nozim Research Institute for Development of Digital Technologies and Artificial Intelligence Laboratory 'Modeling of Complex Systems' Tashkent Uzbekistan

ISBN: (纸本)9781665472296

In this paper we consider a mathematical model and software tool designed for the study of the filtration process a mixture of the 'oil-gas-water' in a porous medium for concrete practical recommendations and administrative decisions. In the course of the numerical calculations were the main parameters of the process and its change ranges, allowing to determine the operation of the reservoir depending on the length and capacity, also the number of wells and production rates. The suitability of the developed algorithm and program for calculation of fields of pressure and saturation are demonstrated, as well as field development indicators in the 'Oil-Gas' systems, the 'oil-water', 'oil-gas-water'. Results of numerical experiments on computer are presented as graphics. © 2022 IEEE.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Assessing Speech-to-Text Translation Quality: An Overview of Key Metrics

Assessing Speech-to-Text Translation Quality: An Overview of...

引用

2024 International Conference on Decision Aid Sciences and Applications, DASA 2024

作者： Labied, Maria Belangour, Abdessamad Banane, Mouad LTIM Hassan II University Laboratory of Information Technology and Modeling Ben m'Sik Faculty of Sciences Casablanca Morocco Hassan II University Laboratory of Artificial Intelligence & Complex Systems Engineering Faculty of Legal Economic and Social Sciences Casablanca Morocco

ISBN: (纸本)9798350369106

Speech-to-text translation plays a pivotal role in numerous real-world applications, from virtual assistants to live translations. Ensuring high-quality translations requires robust evaluation metrics tailored to different model architectures. This paper presents a detailed review of key metrics for evaluating speech-to-text translation quality. Standard metrics like Character Error Rate, Word Error Rate, and BLEU are introduced, with an explanation of their calculation methods. The paper also explores metrics tailored to cascade models, addressing WER for automatic speech recognition, BLEU, METEOR, and Translation Edit Rate for machine translation, while considering pipeline latency in real-time scenarios. For end-to-end models, metrics such as BLEU, ROUGE, METEOR, and Joint Error Rate are discussed, along with the impact of end-to-end latency. The paper concludes by underscoring the importance of selecting the appropriate evaluation metrics and suggests areas for future research in metric development and model assessment. © 2024 IEEE.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Fine-Tuning Whisper for Speech Translation: A Case Study on Translating Darija to Arabic

Fine-Tuning Whisper for Speech Translation: A Case Study on ...

引用

2024 International Conference on Decision Aid Sciences and Applications, DASA 2024

作者： Labied, Maria Belangour, Abdessamad Banane, Mouad Laboratory of Information Technology and Modeling Ben m'Sik Faculty of Sciences LTIM Hassan II University Casablanca Morocco Laboratory of Artificial Intelligence & Complex Systems Engineering Faculty of Legal Economic and Social Sciences Hassan II University Casablanca Morocco

ISBN: (纸本)9798350369106

The task of translating Darija, a dialect of Arabic, into Modern Standard Arabic (MSA) presents significant challenges due to linguistic variations and the lack of representation in major models. Traditional speech recognition and translation systems, like Whisper, are generally trained on more standardized languages and often struggle with dialects like Darija. To address this issue, we fine-tuned the small version of Whisper on the Darija-C corpus for speech-to-text translation into MSA. Our approach involved adapting Whisper's translation capabilities by fine-tuning the model with 5000 training steps and evaluating its performance using the BLEUmetric. The fine-tuned model achieved a BLEU score of 0.65, reflecting a marked improvement over the baseline. This research demonstrates the effectiveness of fine-tuning models for dialectal translation tasks, offering insights into how underrepresented dialects like Darija can be better supported by advanced speech translation systems. © 2024 IEEE.

关键词： Darija-C Fine-Tuning Moroccan Darija Speech Translation Whisper

来源：评论

学校读者我要写书评

暂无评论

An artificial Immune System Algorithm for the Permutation Flow Shop Scheduling Problem with Sequence Dependent Setup Times 10

An Artificial Immune System Algorithm for the Permutation Fl...

引用

10th International Conference on Optimization and Applications, ICOA 2024

作者： Mesmar, Khadija Lebbar, Maria Aqil, Said Asebriy, Zahra Allali, Karam National School of Mines of Rabat Rosdm Research Team Lmaid Laboratory of Applied Mathematics and Business Intelligence Rabat Morocco Ensam Casablanca Hassan Ii University Laboratory Artificial Intelligence and Complex Systems Engineering Casablanca Morocco Est Essaouira Cadi Ayyad University Laboratory mathematics Computer sciences and modeling of complex systems Essaouira Morocco Fst Mohammadia Hassan Ii University Laboratory of Mathematics Computer Science and Applications Mohammadia Morocco

ISBN: (纸本)9798350387353

This paper is concerned with a set of independent jobs in a permutation flow shop that have setup times on a set of machines. The M-IAIS algorithm, which is based on immunoglobulins and has been modified, is designed to reduce the makespan. A unique procedure known as B cell repertoire updating, which comprises size C of jobs that are taken out of the sequence and reinserted into all possible positions of the remaining sequences of jobs, has an impact on the M-IAIS algorithm's performance. To increase the convergence velocity, three iterations of the algorithm are suggested. Thorough testing was done, and the three versions' performances on a set (benchmark) are compared. Problem instances are described in Sistemas de Optimizacion Aplicada (http://***/problem-instances) In the paper, percentage deviation is considered to evaluate the performance of the proposed algorithms. Furthermore, the ensuing outcomes demonstrate its suitability for addressing for the Permutation Flow Shop Scheduling Problem with Sequence Dependent Setup Times. © 2024 IEEE.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Delve deep into End-To-End Automatic Speech Recognition Models

Delve deep into End-To-End Automatic Speech Recognition Mode...

引用

2023 International Seminar on Application for Technology of Information and Communication, iSemantic 2023

作者： Labied, Maria Belangour, Abdessamad Banane, Mouad Hassan Ii University Ben m'Sik Faculty of Sciences Laboratory of Information Technology and Modeling Casablanca Morocco Hassan Ii University Faculty of Legal Economic and Social Sciences Laboratory of Artificial Intelligence & Complex Systems Engineering Casablanca Morocco

ISBN: (纸本)9798350339215

Automatic Speech Recognition (ASR) has experienced significant advancements in recent years, with end-to-end approaches emerging as a promising paradigm shift. Unlike traditional ASR systems that rely on a pipeline of separate components, end-to-end models aim to directly transcribe speech inputs into text using deep learning architectures. In this paper, we conduct a comprehensive study on end-to-end ASR models. We review various architectures employed in end-to-end ASR, including recurrent neural networks (RNNs), convolutional neural networks (CNNs), and transformer-based models. We explore different training methodologies, loss functions, and optimization algorithms used in end-to-end ASR. Additionally, we discuss large-scale datasets commonly used for training and evaluate the performance of end-to-end models using established evaluation metrics such as word error rate (WER). Furthermore, we analyze the strengths and weaknesses of end-to-end ASR models, highlight their applications in real-world scenarios, and discuss open challenges and future directions. By providing this comprehensive study, we aim to facilitate a deeper understanding of end-to-end ASR models and their potential for driving advancements in speech recognition technology. © 2023 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Computer Simulation of Three-Dimensional Gas Filtration Process in Porous Media

Computer Simulation of Three-Dimensional Gas Filtration Proc...

引用

2021 International Conference on Information Science and Communications Technologies, ICISCT 2021

作者： Kurbonov, Nozim Research Institute for Development of Digital Technologies and Artificial Intelligence Laboratory 'Modeling of Complex Systems' Tashkent Uzbekistan

ISBN: (纸本)9781665432580

The paper deals with development of multidimensional model of gas filtration in porous media and numerical algorithm for solving the initial-boundary value problem. The model consists of nonlinear differential equation in partial derivatives with the appropriate initial and boundary conditions. The model makes it possible to evaluate impact of main factors affecting the filtration process such as permeability, porosity, thickness of layers, gas recovery coefficient, viscosity etc. That gives us opportunity to solve practical problems of forecasting the process of gas production as well as making decisions on exploitation of gas deposits. © 2021 IEEE.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

DARIJA-C: Towards a Moroccan DARIJA Speech recognition and speech-to-text Translation Corpus 1

DARIJA-C: Towards a Moroccan DARIJA Speech recognition and s...

引用

1st International Conference in Advanced Innovation on Smart City, ICAISC 2023

作者： Labied, Maria Belangour, Abdessamad Banane, Mouad Laboratory of Information Technology and Modeling LTIM Hassan II University Ben M'Sik Faculty of Sciences Casablanca Morocco Laboratory of Artificial Intelligence & Complex Systems Engineering Faculty of Legal Economic and Social Sciences Hassan II University Casablanca Morocco

ISBN: (纸本)9781665472753

This paper introduces an automated collection of a speech corpus for the Moroccan Arabic dialect 'Darija' (DARIJA-C) which is intended for speech-to-text translation from Moroccan Darija into classical Arabic language. The DARIJA-C corpus is designed for Moroccan Darija automatic speech-to-text translation purposes. Nevertheless, it can be useful for automatic speech recognition of this dialect. To address both scale and sustainability, the DARIJA-C project uses crowdsourcing to collect and validate speech transcriptions and translations. By providing an automatic web platform for recording speech, along with their corresponding translation by distinct unknown speakers. The first versions of the Darija-C dataset will include only the translation of Moroccan Darija speech to classical Arabic. In later versions, we will include the translation of Moroccan Darija into other international languages such as French, and English,... The goal of this work is to build the largest crowdsourced corpus of Darija speech, which to our knowledge will be the first corpus for Moroccan Darija Speech-to-text translation. © 2023 IEEE.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

Annealing approach to root finding

引用

Physical Review E 2024年第2期110卷 025305-025305页

作者： Junghyo Jo Alexandre Wagemakers Vipul Periwal Department of Physics Education Center for Theoretical Physics and Artificial Intelligence Institute School of Computational Sciences Nonlinear Dynamics Chaos and Complex Systems Group Departamento de Física Laboratory of Biological Modeling National Institute of Diabetes and Digestive and Kidney Diseases

The Newton-Raphson method is a fundamental root-finding technique with numerous applications in physics. In this study, we propose a parameterized variant of the Newton-Raphson method, inspired by principles from physics. Through analytical and empirical validation, we demonstrate that this approach offers increased robustness and faster convergence during root-finding iterations. Furthermore, we establish connections to the Adomian series method and provide a natural interpretation within a series framework. Remarkably, the introduced parameter, akin to a temperature variable, enables an annealing approach. This advancement sets the stage for a fresh exploration of numerical iterative root-finding methodologies.

关键词： Numerical techniques Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Speech Translation From Darija to Classical Arabic: Performance Analysis of Whisper, SeamlessM4T, and S2T Models

引用

IEEE Access 2025年 13卷 90527-90544页

作者： Maria Labied Abdessamad Belangour Mouad Banane Laboratory of Information Technology and Modeling (LITM) Faculty of Sciences Ben M’sik Hassan II University Casablanca Morocco Laboratory of Artificial Intelligence and Complex Systems Engineering Faculty of Legal Economic and Social Sciences Hassan II University Casablanca Morocco

This study evaluates the performance of advanced speech-to-text translation models—Whisper large-v3, SeamlessM4T, and S2T—fine-tuned on the Darija-C corpus to translate Darija speech into Classical Arabic text. Darija, a widely spoken Arabic dialect, presents significant challenges for automated translation due to its linguistic complexity and lack of standardized resources. The fine-tuning process adapted each model to the nuances of Darija speech, aiming to optimize translation quality and linguistic alignment with Classical Arabic. The primary goal of this research is to identify the model achieving the highest BLEU score, which serves as a benchmark for translation accuracy and fluency. Comprehensive experiments highlight the comparative strengths and limitations of these models, with Whisper large-v3 emerging as a strong contender alongside SeamlessM4T and S2T. Results reveal the critical role of fine-tuning in improving the performance of pre-trained models on low-resource dialects. This work contributes to the development of effective speech translation systems for Arabic dialects, offering insights into optimizing model architectures for underrepresented languages and enhancing their practical applicability in real-world scenarios.

关键词： Translation Adaptation models Speech to text Biological system modeling Multilingual Decoding Training Accuracy Vocabulary Transformers

来源：评论

学校读者我要写书评

暂无评论

Computer modelling of the fluid filtration process in porous media

Computer modelling of the fluid filtration process in porous...

引用

International Conference on Information Science and Communications Technologies (ICISCT)

作者： Nozim Kurbonov Laboratory ”Modeling of complex systems” Research institute for development of digital technologies and artificial intelligence Tashkent Uzbekistan

In this paper we consider a mathematical model and software tool designed for the study of the filtration process a mixture of the “oil-gas-water” in a porous medium for concrete practical recommendations and administrative decisions. In the course of the numerical calculations were the main parameters of the process and its change ranges, allowing to determine the operation of the reservoir depending on the length and capacity, also the number of wells and production rates. The suitability of the developed algorithm and program for calculation of fields of pressure and saturation are demonstrated, as well as field development indicators in the “Oil-Gas” systems, the “oil-water”, “oil-gas-water”. Results of numerical experiments on computer are presented as graphics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：