Beschreibung
This book includes most of Dr. Zhenhao Ge's PhD work in the area of mispronunciation detection, for both language learning and speech recognition adaptation and improvement. In the applications for language learning, this book introduced one type of context adaptive features, that can be used to improve the accuracy of detection mispronunciations. It also described a method based on Principle Component Analysis, to hierarchically detect the parts of mispronunciations for language learners, in an efficient and accurate way. In the applications for speech recognition adaptation and improvement, this book explained how to improve name recognition by learning acceptable pronunciation variations. It also provided practical approaches based on Gaussian Mixture Modals and Linear Discriminative Analysis for accent detection and classification, in order to make speech recognition adaptive to accents and further improve performance.
Autorenporträt
Dr. Zhenhao Ge graduated from Purdue with PhD in Electrical and Computer Engineering in 2013 and are currently working as Senior Speech Processing Engineer in Interactive Intelligence Inc. His research interests include applying pattern recognition and machine learning techniques for speech applications and natural language processing.
Herstellerkennzeichnung:
OmniScriptum SRL
Str. Armeneasca 28/1, office 1
2012 Chisinau
MD
E-Mail: info@omniscriptum.com
