Discriminative training for neural predictive coding applied to speech features extraction

Author

Chetouani, Mohamed ; Gas, B. ; Zarader, J.L.

Author_Institution

Lab. des Instrum. et Syst. d´Ile de France, Paris VI Univ.

Volume

1

fYear

2002

fDate

6/24/1905 12:00:00 AM

Firstpage

852

Lastpage

857

Abstract

We present a predictive neural network called neural predictive coding (NPC). This model is used for nonlinear discriminant features extraction applied to phoneme recognition. We validate the nonlinear prediction improvement of the NPC model. We also, present a new extension of the NPC model: NPC-3. In order to evaluate the performances of the NPC-3 model, we carried out a study of Darpa-Timit phonemes (in particular /b/, /d/, /g/ and /p/, /t/, /q/ phonemes) recognition. Comparisons with traditional coding methods are presented. We also show how an adaptative constraint allows improvements on the recognition task

Keywords

feature extraction; learning (artificial intelligence); linear predictive coding; neural nets; speech recognition; Darpa-Timit phonemes recognition; NPC-3; adaptative constraint; discriminative training; neural predictive coding; nonlinear discriminant features extraction; phoneme recognition; predictive neural network; speech features extraction; Feature extraction; Instruments; Linear predictive coding; Mel frequency cepstral coefficient; Neural networks; Nonlinear filters; Predictive coding; Predictive models; Speech coding; Speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on

Conference_Location

Honolulu, HI

ISSN

1098-7576

Print_ISBN

0-7803-7278-6

Type

conf

DOI

10.1109/IJCNN.2002.1005585

Filename

1005585