Research on Handwritten Note Recognition in Digital Music Classroom Based on Deep Learning,ERICDATA高等教育知識庫
高等教育出版
熱門: 謝傳崇  羅文君  葉俊廷  紀金山  SDGs  簡淑芸  
高等教育出版
首頁 臺灣期刊   學校系所   學協會   民間出版   大陸/海外期刊   政府機關   學校系所   學協會   民間出版   DOI註冊服務
篇名
Research on Handwritten Note Recognition in Digital Music Classroom Based on Deep Learning
並列篇名
Research on Handwritten Note Recognition in Digital Music Classroom Based on Deep Learning
作者 Yanfang Wang
英文摘要
Music is an indispensable subject in quality education, which plays an important role in improving students’ overall quality. Traditional music teaching is mainly a one-way teaching led by teachers. The teaching style is monotonous and the teaching resources are lacking. It is in sharp contrast with diversified music learning, which affects students’ mastery of basic music skills. The research of this paper is mainly based on the metaphor of “paper and pen”, and the user-centered natural interaction method is used as the design idea to identify the handwritten notes in music teaching in order to provide a natural and efficient teaching environment for music education. Handwritten note recognition draws on the idea of metric learning. Based on the deep Gaussian process model, a non-parametric model, a deep Gaussian matching network for small batch handwritten note recognition is proposed. The framework can adaptively learn a deep structure that can effectively map the labeled support set and unlabeled samples to its label, while avoiding overfitting due to insufficient training data. In the training stage of the deep Gaussian process model, the standardized flow method is used to construct a flexible variational distribution, which improves the quality of inference. Gaussian Processes (GP) are type of supervised learning system that can be used to solve problems like regression and probabilistic classification. Gaussian processes have the following advantages: The forecast generalizes the data from the observations. And when sparse the Gaussian model to reduce the amount of calculation, the optimal k-means method is used to find false points. Experiments were carried out on the handwritten note data set collected in the digital music classroom. The experimental results show that compared with the traditional deep neural network model, the accuracy of the algorithm proposed in this paper has increased from 88% to 94.7% in a single learning sample, and the model proposed in this paper does not rely on fine-tuning and controls the actual calculation amount. The handwritten note recognition effect is better, and it has good application prospects in digital music classrooms.
起訖頁 1443-1455
關鍵詞 Digital music classroomHandwritten note recognitionDeep learningGaussian processNon-parametric estimation
刊名 網際網路技術學刊  
期數 202111 (22:6期)
出版單位 台灣學術網路管理委員會
DOI 10.53106/160792642021112206020  複製DOI
QR Code
該期刊
上一篇
Design and Research of Intelligent Screening System for Graduate Recruitment Based on Big Data Assisted Ontology-based Blockchain Design

高等教育知識庫  閱讀計畫  教育研究月刊  新書優惠  

教師服務
合作出版
期刊徵稿
聯絡高教
高教FB
讀者服務
圖書目錄
教育期刊
訂購服務
活動訊息
數位服務
高等教育知識庫
國際資料庫收錄
投審稿系統
DOI註冊
線上購買
高點網路書店 
元照網路書店
博客來網路書店
教育資源
教育網站
國際教育網站
關於高教
高教簡介
出版授權
合作單位
知識達 知識達 知識達 知識達 知識達 知識達
版權所有‧轉載必究 Copyright2011 高等教育文化事業有限公司  All Rights Reserved
服務信箱:edubook@edubook.com.tw 台北市館前路 26 號 6 樓 Tel:+886-2-23885899 Fax:+886-2-23892500