A CLASSIFIED SPLIT VECTOR QUANTIZATION OF LFS PARAMETERS

Split vector quantization (SVQ) of LSF parameter suffers from huge com plexity and storage requirements. Increasing the number of subvectors M results in a considerable decrease in both complexity and storage, b ut at the expense of rapid degradation in performance as M increases. To alleviate this problem, we propose classified SVQ (CSVQ), which use s class-dependent splitting and bit allocation schemes combined with a classified VQ structure. For practical applications, we designed two CSVQ structures. Experimental results have shown that both of the CSVQ schemes achieve nearly transparent quantization at 28 bits/frame whil e requiring much less complexity than the conventional SVQ. (C) 1997 E lsevier Science B.V.