Title :
Systolic codebook generation [speech recognition]
Author :
Beresford-Smith, Bryan ; Breckling, Jens ; Schröder, Heiko ; Summons, Peter
Author_Institution :
Australian Nat. Univ., Camberra, ACT, Australia
fDate :
4/1/1993 12:00:00 AM
Abstract :
Two systolic codebook generators which are based on finding clusters in a multidimensional space with a suitable dissimilarity measure are presented. They are implemented on two novel VLSI architectures. For segments of speech represented by a sequence of k linear-predictive-coding vectors, the first version finds r large clusters in time O(rk), and the second version finds the maximal r clusters in time O(r log k)
Keywords :
VLSI; linear predictive coding; speech coding; speech recognition; systolic arrays; LPC vectors; VLSI architectures; clusters; dissimilarity measure; linear-predictive-coding vectors; multidimensional space; speech recognition; speech segments; systolic codebook generators; Australia; Extraterrestrial measurements; Hidden Markov models; Iterative algorithms; Linear predictive coding; Multidimensional systems; Speech processing; Speech recognition; Vector quantization; Very large scale integration;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
