Title :
Statistical modeling for dysphonic classification
Author :
Ghelis, Assia ; Guerti, Mhania ; Fredouille, Corinne
Author_Institution :
LEREC Lab., Badji Mokhtar Univ., Annaba, Algeria
Abstract :
The objective of our work is to develop an automatic system to evaluate Arabic/French dysphonic classification by modeling speech signals using a statistical modeling based on a Gaussian Mixture Model (GMM), which is state of art in speaker recognition. Speakers were conducted at Annaba University Hospital center (CHU) in the ENT service in the presence of a group composed of 8 medical specialists. Results of the experiment show that an automatic system is able to identify dysphonic speakers with an acceptable performance either in French or in Arabic language.
Keywords :
Gaussian processes; natural language processing; pattern classification; speaker recognition; statistical analysis; Annaba University Hospital center; Arabic language; Gaussian mixture model; automatic system; dysphonic classification; speaker recognition; speech signals; statistical modeling; Acoustic measurements; Acoustic signal detection; Frequency; Hospitals; Laboratories; Loudspeakers; Natural languages; Pathology; Speaker recognition; Speech analysis; GMM; Speaker recognition; dysphonia;
Conference_Titel :
Design and Technology of Integrated Systems in Nanoscale Era (DTIS), 2010 5th International Conference on
Conference_Location :
Hammamet
Print_ISBN :
978-1-4244-6338-1
DOI :
10.1109/DTIS.2010.5487588
