DocumentCode :
3630079
Title :
Phoneme Recognition as a Member of Predefined Class using Hybrid Cascaded LVQ/Elman Neural Network
Author :
Zikrija Avdagic;Adnan Nuhic;Samim Konjicija
Author_Institution :
Faculty of Electrical Engineering of University of Sarajevo. zikrija.avdagic@etf.unsa.ba
fYear :
2007
Firstpage :
1195
Lastpage :
1198
Abstract :
What is presented is a new approach for implementing Bosnian phoneme recognition. While most of the literature on phoneme recognition is based on hidden Markov models (HMM), or on the recognition by neural networks (NN) of one type, the present system is implemented by hybrid cascaded LVQ/Elman NN. This model was created, because we noted that some types of NN achieve better recognition rate for some phonemes, while the other types of NN better recognize other phonemes. Presented system uses LVQ NN as a front-end recognizer, and depending on the obtained output, makes the re-recognition of the same phoneme by Elman NN. This system achieved higher recognition accuracy then standalone NN models.
Keywords :
"Neural networks","Hidden Markov models","Speech recognition","Speech processing","Finite impulse response filter","Signal processing","Cepstral analysis","Facial animation","Acoustic testing","Automatic speech recognition"
Publisher :
ieee
Conference_Titel :
Signal Processing and Communications, 2007. ICSPC 2007. IEEE International Conference on
Print_ISBN :
978-1-4244-1235-8
Type :
conf
DOI :
10.1109/ICSPC.2007.4728539
Filename :
4728539
Link To Document :
بازگشت