Title :
Analysis and Recognition of Esophageal Speech
Author :
Mantilla, Alfredo ; Perez-Meana, Hector ; TreviNo, Jorge ; Ortal, Patricia
Author_Institution :
Departamento de Electron., ITESM, Mexico City
Abstract :
This paper presents a system for analysis and recognition of voiced segments in Spanish esophageal speech. This algorithm involves voiced speech segments analysis methods based in energy criterion and zero crossing, as well as, feature extraction using a wavelet function based on the inner ear model and the recognition stage through the artificial neural networks. Simulation results are presented for normal and esophageal speech
Keywords :
feature extraction; neural nets; speech recognition; wavelet transforms; Spanish esophageal speech recognition; artificial neural networks; energy criterion; feature extraction; inner ear model; voiced segment recognition; voiced speech segments analysis methods; wavelet function; zero crossing; Artificial neural networks; Esophagus; Humans; Larynx; Signal processing algorithms; Speech analysis; Speech processing; Speech recognition; Speech synthesis; Wavelet transforms; esophageal speech; feature extraction; neural networks; voiced and unvoiced classification; wavelet transform;
Conference_Titel :
Signal Processing and Information Technology, 2006 IEEE International Symposium on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-9753-3
Electronic_ISBN :
0-7803-9754-1
DOI :
10.1109/ISSPIT.2006.270778
