Implementation and analysis of training algorithms for the classification of infant cry with feed-forward neural networks

Author

Orozco, José ; Reyes-García, Carlos A.

Author_Institution

INAOE, Puebla, Mexico

fYear

2003

fDate

4-6 Sept. 2003

Firstpage

271

Lastpage

276

Abstract

We present the development of an automatic recognition system of infant cry, with the objective to classify two types of cry: normal and pathological cry from deaf babies. We used acoustic characteristics obtained by the linear prediction technique and as a classifier a feedforward neural network that was trained with several learning methods, resulting better the scaled conjugate gradient algorithm. Current results are shown, which, up to the moment, are very encouraging with an accuracy up to 94.3%.

Keywords

acoustic analysis; acoustic signal processing; conjugate gradient methods; feature extraction; feedforward neural nets; learning (artificial intelligence); linear predictive coding; pattern classification; acoustic characteristics; acoustic feature extraction; automatic recognition system; feed-forward neural networks; infant cry classification; infant cry recognition; learning methods; linear prediction technique; pathologies detection; scaled conjugate gradient algorithm; training algorithms; Acoustic signal detection; Algorithm design and analysis; Classification algorithms; Deafness; Feedforward neural networks; Feedforward systems; Neural networks; Pain; Pathology; Pediatrics;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Signal Processing, 2003 IEEE International Symposium on

Print_ISBN

0-7803-7864-4

Type

conf

DOI

10.1109/ISP.2003.1275851

Filename

1275851