Data augmentation method based on three-dimensional measurement for silent speech recognition

Reducing the burden of data collection is crucial for advancing speech recognition research. Hence, this research focuses on exploring methods to enhance machine learning from limited data by augmenting the training data based on three-dimensional measurements in the field of Japanese silent speech...

Full description

Saved in:
Bibliographic Details
Published inAcoustical Science and Technology Vol. 45; no. 6; pp. 329 - 332
Main Author Ota, Kenko
Format Journal Article
LanguageEnglish
Published Tokyo ACOUSTICAL SOCIETY OF JAPAN 01.11.2024
Japan Science and Technology Agency
Subjects
Artificial neural networks
Data augmentation
Data collection
Dimensional measurement
Error analysis
Linear transformations
Machine learning
Silent speech recognition
Speech recognition
Three-dimensional measurement
Online AccessGet full text

Cover

More Information
Summary:Reducing the burden of data collection is crucial for advancing speech recognition research. Hence, this research focuses on exploring methods to enhance machine learning from limited data by augmenting the training data based on three-dimensional measurements in the field of Japanese silent speech recognition. We compared the connectionist temporal classification losses during training and the recognition performance with and without key data augmentation techniques to evaluate the effectiveness of the proposed method utilizing the direct linear transformation method. In this case, the deep neural network was trained successfully, resulting in a reduced phoneme error rate.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1346-3969
1347-5177
DOI:10.1250/ast.e24.53