Title :
Stop classification using DESA-1 high resolution formant tracking
Author :
Foote, J.T. ; Mashao, D.J. ; Silverman, H.F.
Author_Institution :
Div. of Eng., Brown Univ., Providence, RI, USA
Abstract :
The results using the DESA-1 (discrete energy separation algorithm-1) quadratic frequency estimator to determine the frequency and rate of change of the second-formant frequency are shown. It is shown for different vowel environments that the DESA-1 algorithm can extract sufficient information to classify stops from vocalic data. The performance is demonstrated to be superior to that of a formant tracker using a more conventional pitch-synchronous LPC (linear predictive coding) analysis. The DESA-1 method is computationally simple enough to be performed in real time.<>
Keywords :
computational complexity; real-time systems; speech recognition; tracking; DESA-1 method; alveolar stops; discrete energy separation algorithm; high resolution formant tracking; labial stops; palatal stops; performance; quadratic frequency estimator; real time; stop classification; vowel environments;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1993. ICASSP-93., 1993 IEEE International Conference on
Conference_Location :
Minneapolis, MN, USA
Print_ISBN :
0-7803-7402-9
DOI :
10.1109/ICASSP.1993.319413
