CONF marios-rr-04-59b/IDIAP LP-TRAP: Linear predictive temporal patterns Athineos, Marios Hermansky, Hynek Ellis, Daniel P. W. EXTERNAL http://publications.idiap.ch/attachments/reports/2004/rr04-59.pdf PUBLIC http://publications.idiap.ch/index.php/publications/showcite/marios-rr-04-59 Related documents 2004 IDIAP RR 04-59 Autoregressive modeling is applied for approximating the temporal evolution of spectral density in critical-band-sized sub-bands of a segment of speech signal. The generalized autocorrelation linear predictive technique allows for a compromise between fitting the peaks and the troughs of the Hilbert envelope of the signal in the sub-band. The cosine transform coefficients of the approximated sub-band envelopes, computed recursively from the all-pole polynomials, are used as inputs to a TRAP-based speech recognition system and are shown to improve recognition accuracy. REPORT marios-rr-04-59/IDIAP LP-TRAP: Linear predictive temporal patterns Athineos, Marios Hermansky, Hynek Ellis, Daniel P. W. EXTERNAL http://publications.idiap.ch/attachments/reports/2004/rr04-59.pdf PUBLIC Idiap-RR-59-2004 2004 IDIAP Accepted for publication in ICSLP 2004 Autoregressive modeling is applied for approximating the temporal evolution of spectral density in critical-band-sized sub-bands of a segment of speech signal. The generalized autocorrelation linear predictive technique allows for a compromise between fitting the peaks and the troughs of the Hilbert envelope of the signal in the sub-band. The cosine transform coefficients of the approximated sub-band envelopes, computed recursively from the all-pole polynomials, are used as inputs to a TRAP-based speech recognition system and are shown to improve recognition accuracy.