Frequency and time filtering of filter-bank energies for HMM speech recognition
Document typeConference lecture
PublisherH. TIMOTHY BRUMMELL, WILLIAM IDSARDI CITATION DELAWARE, NEW CASTLE, DELAWARE
Rights accessRestricted access - publisher's policy
In speech recognition, a discriminative frequency weighting can be achieved by decorrelating the frequency sequence of log mel-scaled filter-bank energies with a computationally inexpensive filter. We show how the spectral parameters that result from this kind of frequency filtering, both alone and combined with filtering of their time trajectories, are competitive with respect to the conventional cepstral representations of speech signals
CitationNadeu, C. [et al.]. Frequency and time filtering of filter-bank energies for HMM speech recognition. A: ICSLP'96. INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING. "Proceedings Fourth International Conference on Spoken Language Processing". Philadelphia: H. TIMOTHY BRUMMELL, WILLIAM IDSARDI CITATION DELAWARE, NEW CASTLE, DELAWARE, 1996, p. 430-433.
|Frequency and t ... HMM speech recognition.pdf||433,1Kb||Restricted access|