Title :
Reducing redundant computation in HMM evaluation
Author :
Deller, J.R., Jr. ; Snider, R.K.
Author_Institution :
Dept. of Electr. Eng., Michigan State Univ., East Lansing, MI, USA
fDate :
10/1/1993 12:00:00 AM
Abstract :
Redundant computations occur when a set of hidden Markov models (HMMs) is evaluated with respect to an observation string. A formal restructuring of the HMM allows the redundancy to be identified and removed. An 𝒪([1-κ]NT) complexity HMM results, with the compression index κ ∈ [0,1) depending upon several factors. Isolated-word recognition experiments illustrate the results
Keywords :
computational complexity; data compression; hidden Markov models; redundancy; speech coding; speech recognition; HMM evaluation; complexity; compression index; hidden Markov models; isolated-word recognition experiments; observation string; redundant computation reduction; Databases; Hidden Markov models; Linear predictive coding; Poles and zeros; Predictive models; Quantization; Reflection; Speech analysis; Speech coding; Speech synthesis;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
