Title :
Transitional speech units and their representation by regressive Markov states: applications to speech recognition
Author :
Deng, L. ; Sameti, H.
Author_Institution :
Dept. of Electr. & Comput. Eng., Waterloo Univ., Ont., Canada
fDate :
7/1/1996 12:00:00 AM
Abstract :
Transitional speech units for American English are proposed and constructed via assimilation of active articulatory features. The effectiveness of a feature-based system exploiting the transitional speech units is demonstrated in evaluation experiments, where the new system with use of quadratic regressive states is shown to achieve error rate reduction of 21% compared with the system using only static subphonemic units
Keywords :
hidden Markov models; speech recognition; statistical analysis; American English; active articulatory features; error rate reduction; evaluation experiments; feature-based system; quadratic regressive states; regressive Markov states; representation; speech recognition; static subphonemic units; transitional speech units; Error analysis; Hidden Markov models; Manipulator dynamics; Polynomials; Speech analysis; Speech processing; Speech recognition; Vocabulary;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
