DocumentCode :
2355839
Title :
Speech pathological study of epinosic patient using artificial neural network
Author :
Zadgaonkar, A.S. ; Shukla, Anupam ; Tiwari, Atul
Author_Institution :
Dept. of Electr. Eng., Gov. Eng. Coll., Raipur, India
fYear :
1995
fDate :
15-18 Feb 1995
Firstpage :
31107
Lastpage :
31472
Abstract :
The use of phonetics in the field of medical study was first reported in 1911 by H. Gutman. Since than things have changed a great deal. Nowadays widely developed computerized equipments are available in phonetics. In the speech utterances of healthy as well as epinosic patients the vowels and voiced consonants may be distinguished easily for pathological studies. In the present study, phonetic information such as vowel amplitude, formant frequency and time duration were extracted from the utterances of healthy and epinosic (mentally and physically disturbed) patients and used as input to a neural network. It was found that an epinosic patient could be easily detected by using the ANN method. It is hoped that proposed ANN method can also be used effectively by speech therapists, as a tool for mass health care for emotionally disturbed as well as epileptic cases
Keywords :
medical diagnostic computing; medical signal processing; neural nets; patient diagnosis; speech processing; emotionally disturbed cases; epileptic cases; epinosic patient; formant frequencies; mass health care tool; pathological studies; phonetic information; speech pathological study; speech therapists; speech utterances; time duration; voiced consonants; vowel amplitude; Acoustic reflection; Anatomical structure; Artificial neural networks; Biological neural networks; Frequency; Loudspeakers; Medical services; Pathology; Pressure measurement; Speech;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society, 1995 and 14th Conference of the Biomedical Engineering Society of India. An International Meeting, Proceedings of the First Regional Conference., IEEE
Conference_Location :
New Delhi
Print_ISBN :
0-7803-2711-X
Type :
conf
DOI :
10.1109/RCEMBS.1995.532989
Filename :
532989
Link To Document :
بازگشت