Fig 8 - uploaded by Fengming Jiao
Content may be subject to copyright.
Flowchart of how the DBN-SVM-based pronunciation classification and error detection model is constructed
Source publication
The learning model and environment are two major constraints on spoken English learning by Chinese learners. The maturity of computer-aided language learning brings a new opportunity to spoken English learners. Based on speech recognition and machine learning, this paper designs a spoken English teaching system, and determines the overall architect...
Contexts in source publication
Context 1
... on this, a pronunciation classification and error detection model was constructed based on the Deep Belief Network and Support Vector Machine (DBN-SVM) [16] to identify learners' pronunciation problems and propose corrective suggestions for them. Figure 8 shows a flowchart of how the DBN-SVM-based pronunciation classification and error detection model is constructed. The DBN-SVM-based pronunciation classification and error detection model first uses the Hidden Markov Model Toolbox to force the alignment of the pronunciation data files collected from the corpus with the reference text to obtain the alignment time information at the sound velocity level, and uses it as a dataset for the pronunciation classification and error detection model [17]. ...
Context 2
... on this, a pronunciation classification and error detection model was constructed based on the Deep Belief Network and Support Vector Machine (DBN-SVM) [16] to identify learners' pronunciation problems and propose corrective suggestions for them. Figure 8 shows a flowchart of how the DBN-SVM-based pronunciation classification and error detection model is constructed. The DBN-SVM-based pronunciation classification and error detection model first uses the Hidden Markov Model Toolbox to force the alignment of the pronunciation data files collected from the corpus with the reference text to obtain the alignment time information at the sound velocity level, and uses it as a dataset for the pronunciation classification and error detection model [17]. ...
Citations
... AI-powered applications can be applied pronunciation training, making it more engaging for learners (Chuyen et al., 2021). Integrating AI speech recognition and machine learning into teaching systems can personalize pronunciation training and track progress (Jiao et al., 2021). With these advantages, the wider application of AI technology in learning English pronunciation promises to help learners progress in their studies. ...
Artificial intelligence (AI) is taking over many areas of human life, including teaching English pronunciation. Reading Progress is an AI-powered tool in Microsoft Teams developed by Microsoft to assist language learners in practicing pronunciation. However, there is a scarcity of research on this application. This study aims to investigate students’ evaluation on the use of the tool in learning English pronunciation. This research uses an online survey with 123 students when studying various English modules in a university in northern Vietnam. Research results show that Reading Progress is positively evaluated and considered a useful tool to help learners practice pronunciation. This result contributes to providing more research information on the application of Reading Progress in the context of teaching English pronunciation in Vietnam as well as around the world. Thereby, learners and English teachers can refer to and apply in practice, increasing the effectiveness of learning and teaching.
... Thus, pronunciation training and the subsequent evaluation through AI-powered tools are most frequently carried out using commercially available learning applications and programs. Non-specified systems are being tested for accuracy by a non-defined group of speakers (Jiao et al., 2021;Shufang, 2021;Cheng & Wang, 2022) and indicate future improvements of the tools. ...
Artificial intelligence (AI) is taking over many spheres of human life, including language pedagogy. While some areas need to find their ground with AI and resolve ethical issues arising from its use, other spheres of education, such as pronunciation, may benefit from the system’s ability to communicate with learners and provide them with implicit feedback while carrying out different communication tasks. This technical advancement of AI then opens doors for further educational opportunities that have not existed in the past. The study aims to explore the existing use of AI-powered tools in foreign language pronunciation training by meta-analysis of 15 research papers benefiting from using AI and AI-powered tools (mobile and web applications, chatbots, intelligent virtual assistants) and provide suggestions for their future applications in educational practice. The research results also indicate that this field of study is still underrepresented in language pedagogy. The existing experience with AI-powered tools confirms a relatively good experience in developing intelligibility, increasing motivation and addressing the speaking anxiety of foreign language learners in formal and informal learning settings.
... Now, students can study anytime and anywhere through the online learning platform, no longer limited by time and space [1,2]. The online education platform not only breaks through the time and space restrictions, and provides students with rich learning resources, but also provides students with personalized learning recommendation and evaluation [3] through the intelligent learning system. At present, the online teaching model has attracted the attention of many educators at home and abroad, and many universities have deployed and implemented their own learning platforms. ...
Internet + Education makes online teaching gradually penetrate the education industry, and makes the industry enter a great revolution based on information technology. The traditional student learning evaluation system cannot satisfy the actual demand of current learning evaluation. This paper constructs an evaluation model for the foreign language learning of online students. Firstly, the DBSCAN algorithm with distance optimization is used to conduct cluster analysis on the description indicators of student behavior, and the student groups with different behavior characteristics are obtained. Then the ANOVA F-test was used to extract the features of different student groups. Finally, a novel N-Adaboost algorithm based on multiple classifiers is proposed and a model is constructed to evaluate students’ foreign language learning. The experimental results show that the accuracy of the evaluation model is 74.02% in the pass and fail groups and 73.74% in the excellent and non-excellent groups. Students’ listening, speaking, and reading abilities are in a state of upward development overall through the online teaching collaboration platform, but their writing ability is obviously declining. There is a great improvement in foreign language vocabulary. This study provides a new perspective of thinking for the improvement of the quality of school teaching management, the analysis of students’ behavior, and the evaluation of learning situations, and provides a new solution for the problem of students’ learning situations in modern information teaching.
... This section highlights the outcomes of the framework, which is built on WN technology and AI educational robots. Technologies like "Machine Learning (ML)" [31], "Deep Learning (DL)" [32], and "Virtual Reality (VR)" [33] each have unique properties that they apply to. This study compares these strategies with the suggested methodology CapsNet in order to contrast and choose the best technology. ...
Introduction: The use of AI in education can give students a more engaging learning environment and boost their motivation, and it also represents a continuation of research into the problem of human individuality in the modern era. Objectives: This paper examines the challenge of human individuality in Artificial intelligence with the Capsule network (CapsNet) scheme from two vantage points: the practical need to address issues that have arisen with the latest wave of AI advancements and a philosophical examination of how AI has already been put to use in a variety of industries. Methodology: This article investigates the new Internet spoken English teaching method, describing its benefits and providing solutions to its drawbacks, and it describes in detail how wireless technology will be implemented into online spoken English teaching. The technology provides visual representations of each stage of the gesture recognition process to aid learning. The interactive interface guides students through the gesture recognition system using computer vision applications, allowing them to encounter it firsthand; then, the sophisticated and abstract action recognition method is described with a representational illustration, which is helpful for students in elementary and secondary school to gain a more thorough understanding of and develop their capacity for logical reasoning. This will benefit students at elementary and secondary levels because it will help them think more critically and thoroughly. As a final step, we devise an experiment to compare the results of using our CapsNet method to acquire AI knowledge with those of more conventional learning strategies. Results: Experimental findings were analyzed to demonstrate that this approach is useful for acquiring CapsNet and AI and that it increases users' motivation to study and their practical competence.
... Currently, English teaching evaluation methods in colleges and universities include fuzzy comprehensive evaluation method [6], decision tree [7], support vector machine [8], shallow neural network [9], and deep learning methods [10]. Literature [11] uses the method of decision tree and neural network to construct an English teaching evaluation model, and proposes the potential causal relationship of each evaluation index; Literature [12] combines the convolutional neural network and the long and short-term memory network to build an online education evaluation method; Literature [13] proposes a distance teaching evaluation method for colleges and universities based on the decision tree algorithm, which improves the accuracy of the evaluation of the quality of teaching; Literature [14] constructs a method of distance teaching quality evaluation model and achieved better evaluation results; literature [15] used genetic optimization algorithm to improve neural network to complete teaching quality evaluation method, which has faster evaluation speed and accuracy; literature [16] used bottle sea squirt algorithm to optimize the hyperparameters of the support vector machine to construct the public teaching quality evaluation method of the university, which improves the evaluation accuracy; literature [17] improved the evaluation accuracy by dividing and extracting English teaching quality evaluation indexes, establishing the English teaching quality evaluation model based on krill swarm optimization algorithm optimization to enhance the kernel-limit learning machine, which provides new ideas for English teaching evaluation. For the analysis of the above literature, the existing English teaching evaluation methods have the following defects: 1) the evaluation indexes are not comprehensive enough; 2) the teaching quality evaluation method is single, not scientific and diversified; 3) the assessment is only from the perspective of teachers, which is limited [18]. ...
INTRODUCTION: The construction of English evaluation methods in colleges and universities, as the essential part of English teaching in colleges and universities, is conducive to the improvement of the quality of English teaching in colleges and universities, which makes the existing English teaching more objective and reasonable, and makes the means of English teaching rich in science. OBJECTIVES: Aiming at the current wisdom teaching evaluation design methods exist evaluation indexes exist objectivity is not strong, accuracy is poor, single method and other problems. METHODS:Proposes a college English teaching evaluation method based on a deep learning network. First, the evaluation index system of English in colleges and universities is constructed by analyzing the principle of selecting evaluation indexes of English in colleges and universities; then, the deep learning network is improved through self-coder and integrated learning methods to construct the evaluation model of English teaching in colleges and universities; finally, the effectiveness and efficiency of the proposed method is verified through simulation experiment analysis. RESULTS: The results show that the proposed method improves the accuracy of the evaluation model. CONCLUSION: Solved the problems of low evaluation accuracy and non-objective system indexes of English teaching evaluation methods in colleges and universities.
... -Speech Recognition : Here, the concerned models able to convert spoken language into text, for enabling voice assistants and transcription services (Jiao et al., 2021). ...
This study focuses on spam detection within Airtel, Vodacom, and Orange networks,
emphasizing challenges faced by users, including threats and financial losses. Solutions
proposed involve user education, a predictive model, and user-friendly interfaces using
HTML, CSS, and JavaScript. Python and libraries (Pandas, NumPy, etc.) handle ma-
chine learning models trained on Swahili, French, and English datasets with and end-
point(application interface). The study introduces a server interaction system for devel-
opers and enhances monitoring for network operators. The outcomes include a reduction
in spam-related issues, improved network efficiency, and increased customer satisfaction.
Keywords: Spam Detection, Network Operators,Machine Learning Model, application
interface, Programming Languages, Data Analysis, Server Interaction System.
... This section highlights the outcomes of the framework, which is built on WN technology and AI educational robots. Technologies like "Machine Learning (ML)" [31], "Deep Learning (DL)" [32], and "Virtual Reality (VR)" [33] each have unique properties that they apply to. This study compares these strategies with the suggested methodology CapsNet in order to contrast and choose the best technology. ...
Introduction: The use of AI in education can give students a more engaging learning environment and boost their motivation, and it also represents a continuation of research into the problem of human individuality in the modern era.Objectives: This paper examines the challenge of human individuality in Artificial intelligence with the Capsule network (CapsNet) scheme from two vantage points: the practical need to address issues that have arisen with the latest wave of AI advancements and a philosophical examination of how AI has already been put to use in a variety of industries.Methodology: This article investigates the new Internet spoken English teaching method, describing its benefits and providing solutions to its drawbacks, and it describes in detail how wireless technology will be implemented into online spoken English teaching. The technology provides visual representations of each stage of the gesture recognition process to aid learning. The interactive interface guides students through the gesture recognition system using computer vision applications, allowing them to encounter it firsthand; then, the sophisticated and abstract action recognition method is described with a representational illustration, which is helpful for students in elementary and secondary school to gain a more thorough understanding of and develop their capacity for logical reasoning. This will benefit students at elementary and secondary levels because it will help them think more critically and thoroughly. As a final step, we devise an experiment to compare the results of using our CapsNet method to acquire AI knowledge with those of more conventional learning strategies. Results: Experimental findings were analyzed to demonstrate that this approach is useful for acquiring CapsNet and AI and that it increases users' motivation to study and their practical competence.
... Pitch is s a crucial component of human voice and widely recognized as perceptual fundamental of sound that is strongly attached to frequency and can be related to the vocal cord's vibration fundamental frequency, permitting audio frequency recognition. It is among the most essential auditory features of sounds, as well as quality and loudness [23], [30]. ...
... This method uses domain correlation feature transformation to transform and generalize the extracted short-term spectral structure features and envelope features of DES; Different time granularities are used to divide DES into multiple levels, and SVM (Support Vector Machine) classifier is used to extract and recognize the features of DES after multi-level division. For example, Jiao [6] used SVM model in his study to classify and detect the errors in English pronunciation. Deng [7] proposed a method for DES recognition on cloud platform network based on deep neural network. ...
Although there are various recognition methods for digital English speech (DES) on cloud platform network, the accuracy of them is low and the average time consumption is too long. Due to above shortages, this study put forward a self-adaption recognition method based on wavelet neural network (WNN). At first, the study used zero-crossing rate as feature of voiceless and voiced sound of the DES and used WNN to carry out conversion for the English speech information. Then, the study obtained wavelet-scaling function and parameterized wavelet function via zero-pole model. The parameterized wavelet function was used as the feature vector of the recognition. Wavelet function base of DES feature was generated via stretching and translating transformation. The transformed wavelet function was used to build WNN model. Moreover, error entropy function of the recognition was calculated by introducing momentum factor and partial derivative of the error entropy function to adjust parameter of the built WNN model. Thus, the study achieved the self-adaption recognition of DES. Simulation results show that the proposed method has high recognition accuracy and short recognition time.
... Reference [5] proposes a human-machine dialogue system to increase learners' interest and accuracy in oral English practice. Reference [6] introduces a deep belief network to recognize the pronunciation of spoken language and judge whether it is correct or not. Reference [7] analyzes the difficulties existing in oral English teaching through data analysis tools and gives corresponding solutions. ...
The progress of global economic integration has forced English learners to have an urgent need to improve their oral English. College students’ oral English ability is currently the worst of the four abilities of listening, speaking, reading, and writing. The main reasons are internal and external. The internal reason is that the pronunciation characteristics of Chinese students are different from those of English. The external cause is that the practice environment and tools of oral English are not ideal, which affects the improvement of learners’ oral English. This study proposes using a deep learning algorithm (DLA) English in the evaluation of oral English quality to improve learners’ oral English level. The quality of oral English can be comprehensively evaluated in terms of pitch, speed of sound, and rhythm. The standard of pronunciation is the foundation of oral English and is the most critical factor. In many DLAs, the input unit of DNN at a certain moment and its upper and lower moment input units have no relationship and are independent of each other, and the timing dependencies of adjacent units are not fully considered. The results are generally not very good on speech recognition tasks. This study proposes a time-delay neural network (TDNN) and a long short-term memory (LSTM) network to calculate the posterior probability of the model state to model context-dependent features in order to solve this problem. The fusion model TDNN-LSTM is applied in the English spoken pronunciation recognition task. To compare the accuracy of oral English pronunciation, several classic DLAs are introduced. The experimental results show that the method described in this study has a number of advantages. Although the performance improvement of this method in terms of recognition accuracy is not large, a certain degree of improvement is also very important for the oral English teaching assistant system.
