Title :
Cause Estimation of Younger Babies´ Cries from the Frequency Analyses of the Voice - Classification of Hunger, Sleepiness, and Discomfort -
Author :
Mima, Yuki ; Arakawa, Kaoru
Author_Institution :
Dept. of Comput. Sci., Meiji Univ., Kawasaki
Abstract :
A method for estimating why a baby is crying is proposed using frequency analyses of the voice. In this paper, three causes, hunger, sleepiness and discomfort, are considered for classification. Moreover, the method intends for younger babies aged from 2 to 4 months old. Some typical characteristics are obtained for hunger, sleepiness, and discomfort, respectively, in the shape of the power spectra of the crying. By using the difference between these characteristics, a rule-based system is proposed, which can classify cries of babies from 2 to 4 months old in a hungry state, a sleepy state, and a discomfort one. Computer simulations show that about 85% of the babies´ cries are correctly classified
Keywords :
knowledge based systems; speech processing; cause estimation; discomfort classification; hunger classification; power spectra; rule-based system; sleepiness classification; voice frequency analyses; Aging; Computer simulation; Frequency estimation; Knowledge based systems; Pediatrics; Shape; Signal analysis; Signal processing; Speech analysis; State estimation;
Conference_Titel :
Intelligent Signal Processing and Communications, 2006. ISPACS '06. International Symposium on
Conference_Location :
Yonago
Print_ISBN :
0-7803-9732-0
Electronic_ISBN :
0-7803-9733-9
DOI :
10.1109/ISPACS.2006.364828