检索结果-内蒙古大学图书馆

Exploring data augmentation for Amazigh speech recognition with convolutional neural networks

International Journal of Speech Technology 2025年第1期28卷 53-65页

作者： Boulal, Hossam Bouroumane, Farida Hamidi, Mohamed Barkani, Jamal Abarkan, Mustapha LSI Laboratory FP Taza USMBA University Taza Morocco Team of Modeling and Scientific Computing FPN UMP Nador Morocco

In the field of speech recognition, enhancing accuracy is paramount for diverse linguistic communities. Our study addresses this necessity, focusing on improving Amazigh speech recognition through the implementation of three distinct data augmentation methods: Audio Augmentation, FilterBank Augmentation, and SpecAugment. Leveraging Convolutional Neural Networks (CNNs) for speech recognition, we utilize Mel Spectrograms extracted from audio files. The study specifically targets the recognition of the initial ten Amazigh digits. We conducted experiments with a speaker-independent approach involving 42 participants. A total of 27 experiments were conducted, utilizing both original and augmented data. Among the different CNN models employed, the VGG19 model showcased significant promise. Our results demonstrate a maximum accuracy of 95.66%. Furthermore, the most notable improvement achieved through data augmentation was 4.67%. These findings signify a substantial enhancement in speech recognition accuracy, indicating the efficacy of the proposed methods. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing amazigh ASR through convolutional neural networks and MFCC

引用

Multimedia Tools and Applications 2024年 1-20页

作者： Boulal, Hossam Hamidi, Mohamed Barkani, Jamal Abarkan, Mustapha LSI Laboratory FP Taza USMBA University Taza Morocco Team of modeling and scientific computing FPN UMP Nador Morocco

In this study, we developed a speech recognition system for the Amazigh language, specifically targeting the recognition of the initial ten numbers. The system employs four Convolutional Neural Network (CNN) models, including three custom-designed models and a pre-trained VGG19 model. Our experiments utilized a dataset comprising 4200 audio files recorded by 42 distinct speakers, with input features extracted as Mel Frequency Cepstral Coefficients (MFCCs). We tested three normalization methods: no normalization, Cepstral Mean and Variance Normalization (CMVN), and Min-Max normalization. While CMVN generally provided effective standardization, We achieved the highest accuracy of 97.56% using Min-Max normalization with a specific filter size in the third custom CNN model. The VGG19 model, however, showed suboptimal performance. These findings underscore the significance of selecting suitable normalization techniques and model architectures for enhancing speech recognition accuracy. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Amazigh CNN speech recognition system based on Mel spectrogram feature extraction method

引用

International Journal of Speech Technology 2024年第1期27卷 287-296页

作者： Boulal, Hossam Hamidi, Mohamed Abarkan, Mustapha Barkani, Jamal LSI Laboratory FP Taza USMBA University Taza Morocco Team of Modeling and Scientific Computing FPN UMP Nador Morocco

The field of speech recognition makes it simpler for humans and machines to engage with speech. Number-oriented communication, such as using a registration code, mobile number, score, or account number, can benefit from speech recognition for digits. This paper presents our Amazigh automatic speech recognition (ASR) experience based on the deep learning approach. The convolutional neural network (CNN) and Mel spectrogram are exploited to evaluate audio samples and produce spectrograms as a part of the deep learning strategy. To attempt the recognition of the Amazigh numerals, we use a database that includes digits ranging from zero to nine collected from 42 native speakers in total, men and women between the ages of 20 and 40. Our experimental results show that spoken digits in Amazigh can be identified with a maximum accuracy of 93.62%, 94% Precision, and 94% Recall. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Data Augmentation for Amazigh Speech Recognition Using Filter Banks 4th

Data Augmentation for Amazigh Speech Recognition Using Filte...

引用

4th International Conference on Electronic Engineering and Renewable Energy Systems, ICEERE 2024

作者： Boulal, Hossam Hamidi, Mohamed Barkani, Jamal Abarkan, Mustapha Sidi Mohamed Ben Abdellah University of Fez Multidisciplinary Faculty of Taza LSI Laboratory Taza Morocco Mohammed First University of Oujda Multidisciplinary Faculty of Nador Team of Modeling and Scientific Computing Nador Morocco

ISBN: (纸本)9789819606436

Our research aims to enhance the accuracy of Amazigh speech recognition through a data augmentation method. We employ Convolutional Neural Networks (CNNs) for speech recognition, utilizing Mel Spectrograms extracted from audio files. This study focuses on recognizing the first ten Amazigh digits. Our approach centers on refining recognition accuracy and augmenting data by modifying the number of bands in the filter bank. Additionally, our study involves 42 speakers who have participated, applying a speaker-independent approach throughout the experiments. We conducted three experiments using original data and three additional experiments using a combination of original and augmented data. The augmentation method involved changing the number of bands in the filter bank. Through these experiments with different CNN models, one model exhibited a 2.89% increase in accuracy, contributing to the overall improvement in speech recognition accuracy, with the highest achieved level reaching 95.88%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Serious Games Intended to Be Used in Education: Systematic Review 7

Evaluation of Serious Games Intended to Be Used in Education...

引用

7th International Conference on Advanced Communication Technologies and Networking, CommNet 2024

作者： Bouroumane, Farida Hamdaoui, Ikram Abarkan, Mustapha University Mohammed First LaMAO Laboratory Modeling and Scientific Computing Team Multidisciplinary Faculty of Nador Oujda Morocco Sidi Mohamed Ben Abdellah University Laboratory of Engineering Sciences Faculty Polydisciplinaire of Taza Fez Morocco

ISBN: (纸本)9798350367027

Serious games have been used as an innovative teaching strategy to make learning more effective. However, it remains important to identify the serious games that are best suited to the needs, expectations, and pedagogical objectives of a teacher. Moreover, using games in educational situations requires teachers to question the elements that make up the pedagogical and game aspects, universe functioning, and conditions of use. Precise and comprehensive information about the relevance of serious game content to specific learning situations can be obtained by examining the game's components. The aim of this review is to identify, describe, classify and summarize the main elements of existing research on the evaluation of serious games for education. In addition, our study could help identify and analyze potential gaps in the use of serious games in education. A qualitative analysis was conducted by selecting 37 key articles from Google Scholar, Scopus, Web of Science, and PubMed databases. This study could be useful for researchers and practitioners interested in the use of serious games in education. It could also contribute to the continuous improvement of digital learning in education. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Amazigh speech recognition based on the Kaldi ASR toolkit

引用

International Journal of Information Technology (Singapore) 2023年第7期15卷 3533-3540页

作者： Barkani, Fatima Hamidi, Mohamed Laaidi, Naouar Zealouk, Ouissam Satori, Hassan Satori, Khalid Laboratory of Computer Science Signals Automation and Cognitivism Faculty of Sciences Dhar Mahraz University Sidi Mohamed Ben Abdellah Fez Morocco Team of modeling and scientific computing Pluridisciplinary Faculty of Nador Mohammed First University Oujda Morocco

In this work, we offer a new approach to integrating the Amazigh language, which is a less-resourced language, into an isolated speech recognition system by exploiting the Kaldi open-source platform. Our designed system is able to recognize the ten first Amazigh digits and ten daily must-used Amazigh isolated words, which present typical syllabic structure and which are considered a good representative sample of the Amazigh language. The designed speech system was implemented using Hidden Markov Models (HMMs) with different number of Gaussian distributions. In addition, we evaluated our created system performance by varying the feature extraction methods in order to determine the optimal method for maximum performance. The best-obtained result is 93.96% was obtained with Mel Frequency Cepstral Coefficients (MFCCs) technique. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.

关键词： Amazigh language CMU Sphinx4 FBANK GMM HMM Kaldi MFCC PLP Speech recognition

来源：评论

学校读者我要写书评

暂无评论

ON THE FUNDAMENTAL EIGENVALUE GAP OF STURM-LIOUVILLE OPERATORS

arXiv

引用

arXiv 2024年

作者： Ahrami, Mohammed El Allali, Zakaria Harrell II, Evans M. Team of Modeling and Scientific Computing Department of Mathematics Multidisciplinary Faculty of Nador University of Mohammed First Morocco School of Mathematics Georgia Institute of Technology AtlantaGA30332-0160 United States

We use methods of direct optimization as in [9] to find the minimizers of the fundamental gap of Sturm-Liouville operators on an interval, under the constraint that the potential is of single-well form and that the weight function is of single-barrier form, and under similar constraints expressed in terms of *** Codes 34L15, 34B27, 35J60, 35B05 © 2024, CC BY-NC-SA.

关键词： Eigenvalues and eigenfunctions

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Serious Games Intended to Be Used in Education: Systematic Review

Evaluation of Serious Games Intended to Be Used in Education...

引用

International Conference on Advanced Communication Technologies and Networking (CommNet)

作者： Farida Bouroumane Ikram Hamdaoui Mustapha Abarkan LaMAO Laboratory Modeling and Scientific Computing Team Multidisciplinary Faculty of Nador University Mohammed First Oujda Morocco Laboratory of Engineering Sciences Faculty Polydisciplinaire of Taza Sidi Mohamed Ben Abdellah University Fez Morocco

ISBN: (数字)9798350367027

ISBN: (纸本)9798350367034

Serious games have been used as an innovative teaching strategy to make learning more effective. However, it remains important to identify the serious games that are best suited to the needs, expectations, and pedagogical objectives of a teacher. Moreover, using games in educational situations requires teachers to question the elements that make up the pedagogical and game aspects, universe functioning, and conditions of use. Precise and comprehensive information about the relevance of serious game content to specific learning situations can be obtained by examining the game’s components. The aim of this review is to identify, describe, classify and summarize the main elements of existing research on the evaluation of serious games for education. In addition, our study could help identify and analyze potential gaps in the use of serious games in education. A qualitative analysis was conducted by selecting 37 key articles from Google Scholar, Scopus, Web of Science, and PubMed databases. This study could be useful for researchers and practitioners interested in the use of serious games in education. It could also contribute to the continuous improvement of digital learning in education.

关键词： Training Systematics Databases Education Games Serious games Internet Information and communication technology Object recognition Guidelines

来源：评论

学校读者我要写书评

暂无评论

Asymptotic Error Expansion of the Iterated Superconvergent Nyström Method for Urysohn Integral Equations

引用

International Journal of Applied and Computational Mathematics 2021年第6期7卷 1-12页

作者： Allouch, C. Medbouhi, A. Sbibih, D. Tahrichi, M. FPN Team of Modeling and Scientific Computing University Mohammed I Nador Morocco FSO LIABM Laboratory University Mohammed I Oujda Morocco FSO LANO Laboratory University Mohammed I Oujda Morocco EST ANAA Team LANO Laboratory University Mohammed I Oujda Morocco

For a nonlinear Urysohn integral equation with smooth kernel, a superconvergent Nyström method proposed recently in Allouch et al. (BIT Numer Math 258:10777–10790, 2017), based on an interpolatory projection onto a space of piecewise polynomials of degree ≤ r, is shown to have convergence of order 4r for its iterated version. In this paper, a general theorem dealing with asymptotic error expansion for the iterated superconvergent Nyström solution is proved. The Richardson extrapolation is then applied to obtain a more accurate solution of order 4 r+ 2. Numerical results are given to illustrate this improvement of the convergence order. © 2021, The Author(s), under exclusive licence to Springer Nature India Private Limited.

关键词： Gauss points Richardson extrapolation Superconvergent Nyström method Urysohn integral equations

来源：评论

学校读者我要写书评

暂无评论

OPTIMIZING THE FUNDAMENTAL EIGENVALUE GAP OF QUANTUM GRAPHS

arXiv

引用

arXiv 2024年

作者： Ahrami, Mohammed El Allali, Zakaria Harrell, Evans M. Kennedy, James B. Modeling and Scientific Computing Team Multidisciplinary Faculty of Nador Mohammed I University Morocco School of Mathematics Georgia Institute of Technology AtlantaGA30332-0160 United States Grupo de Física Matemática Departamento de Matemática Faculdade de Ciências Universidade de Lisboa Campo Grande Lisboa1749-016 Portugal

We study the problem of minimizing or maximizing the fundamental spectral gap of Schrödinger operators on metric graphs with either a convex potential or a "single-well" potential on an appropriate specified subset. (In the case of metric trees, such a subset can be the entire graph.) In the convex case we find that the minimizing and maximizing potentials are piecewise linear with only a finite number of points of non-smoothness, but give examples showing that the optimal potentials need not be constant. This is a significant departure from the usual scenarios on intervals and domains where the constant potential is typically minimizing. In the single-well case we show that the optimal potentials are piecewise constant with a finite number of jumps, and in both cases give an explicit estimate on the number of points of non-smoothness, respectively jumps, the minimizing potential can have. Furthermore, we show that, unlike on domains, it is not generally possible to find nontrivial bounds on the fundamental gap in terms of the diameter of the graph alone, within the given *** Codes 34B45, 34L15, 35J05, 81Q35 © 2024, CC BY-NC-SA.

关键词： Eigenvalues and eigenfunctions

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：