LP-TRAP: Linear predictive temporal patterns
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.
- URL: http://publications.idiap.ch/downloads/reports/2004/rr04-59.pdf
- Related documents: http://publications.idiap.ch/index.php/publications/showcite/marios-rr-04-59
IDIAP RR 04-59
Record created on 2006-03-10, modified on 2016-08-08