Education Theory: Teaching and LearningOpen Access
Education Theory: Teaching and Learning (ETTL) is an international, peer-reviewed journal dedicated to publishing original, high-quality research across all areas related to education theory, teaching, and learning. The journal provides open-access content for a global readership, aiming to foster scientific research and academic exchange among scholars. ETTL focuses on advancing the development of educational theory and practice, highlighting innovative research from around the world, and promoting international collaboration and knowledge sharing.
ISSN: N/A (Print)
Frequency: Quarterly
ISSN: 2771-9030 (Online)
Website: https://doi.org/10.55571/ettl
Email: journal@4dpublishinggroup.com
Indexing and Abstracting: Google Scholar; CrossRef; ResearchBib; WorldCat; CiteFactor
Email: journal@4dpublishinggroup.com
Indexing and Abstracting: Google Scholar; CrossRef; ResearchBib; WorldCat; CiteFactor
Speech Emotion Recognition of Minnanese Based on Deep Learning
DOI: https://doi.org/10.55571/ettl.2022.06020
Authors: Xin-hua Guo and Miao Zhang
Affiliation: Software College, Quanzhou University of Information Engineering, Quanzhou 362000, China
Information: Education Theory: Teaching and Learning, June 2022 Vol.1, No.3, pp.99-107
Authors: Xin-hua Guo and Miao Zhang
Affiliation: Software College, Quanzhou University of Information Engineering, Quanzhou 362000, China
Information: Education Theory: Teaching and Learning, June 2022 Vol.1, No.3, pp.99-107
Abstract: Minnanese is a representative branch of Fujian dialect.Due to objective reasons such as the late start of the research on emotion recognition of Minnanese speech and the immaturity of technology, there are few research achievements in this area, especially in the area of corpus resources of Minnanese language. Based on the establishment of Minnanese language emotion corpus, soft-Max is adopted as the algorithm of emotion recognition, and emotion recognition is carried out based on CNN model and BLSTM model. In view of the global and temporal characteristics of Minnanese language corpus, an algorithm model based on CNN-BLSTM fusion feature is proposed.Experimental results show that CNN-MOBLSTM fusion feature model is superior to CNN model and BLSTM model in recognition of Minnanese emotion, and it is an effective algorithm model.
Keywords: Deep Learning, Minnanese, Emotion recognition, CNN-MOBLSTME
Cite This Article: Gao X.H and Zhang M. (2022). Speech Emotion Recognition of Minnanese Based on Deep Learning. Education Theory: Teaching and Learning, June 2022 Vol.1, No.3, pp.99-107
Keywords: Deep Learning, Minnanese, Emotion recognition, CNN-MOBLSTME
Cite This Article: Gao X.H and Zhang M. (2022). Speech Emotion Recognition of Minnanese Based on Deep Learning. Education Theory: Teaching and Learning, June 2022 Vol.1, No.3, pp.99-107
