Title :
An artificial neural network-based isolated word speech recognition system for the Romanian language
Author :
Mihaela Oprea;Daniela Şchiopu
Author_Institution :
Petroleum-Gas University of Ploiesti, Department of Automatic Control, Computers and Electronics, Blvd. Bucuresti 39, Ploiesti, 100680, Romania
Abstract :
Automatic speech recognition is a very difficult task to solve due to many factors that influence it (e.g. speaker pronounciation, noisy environment). The paper presents a Romanian isolated word speech recognition system based on feed forward artificial neural networks (ANN). The system was experimented for a set of vocal commands applied to control electronic devices.
Keywords :
"Speech","Feature extraction","Artificial neural networks","Automatic speech recognition","Cepstral analysis"
Conference_Titel :
System Theory, Control and Computing (ICSTCC), 2012 16th International Conference on
Print_ISBN :
978-1-4673-4534-7
