Optimized Subvector Processing in Split Vector Quantization

M.A Ramírez

doi:10.14209/jcis.2010.3

Abstract

Split vector quantization (SVQ) is efficient but suboptimal. Here a renormalization process is proposed for intraframe splitting and joining of subvectors, which integrates gracefully with trained interframe prediction. Renormalization increases the availability of codevectors for the quantization of each subvector in ordered vectors such as the line spectral frequency (LSF) vectors. For 16-dimensional LSF vectors from wideband speech, renormalized SVQ (RSVQ) is shown to achieve a savings of 4 bit/frame over standard SVQ, reaching transparent coding at 42 bit/frame. Further, predictive RSVQ saves an additional 4 bit/frame for transparent coding down to 38 bit/frame.

Highlights

S PLIT vector quantization overcomes the curse of dimensionality inherent in vector quantization by splitting the vector into lower-dimensional subvectors
An amount of suboptimality remains that is referred to as the split loss [2]. It may be partially counteracted by classified vector quantization (VQ) [3] and by combining split VQ with multistage VQ [4]
For split VQ, line spectral frequency (LSF) vector f = {fi}pi=1 is partitioned as f = φT1 φT2 · · · φTς T, where ς is the number of partitions or splits and the ith subvector consists of φi = fδi fδi+1 · · · fδi+Di−1 T

Summary

INTRODUCTION

S PLIT vector quantization overcomes the curse of dimensionality inherent in vector quantization by splitting the vector into lower-dimensional subvectors. It is efficient when the distortion measure used for quantization is separable. It is widely used for the quantization of line spectral frequency (LSF) vectors that represent the shortterm spectral envelope of speech signals [1]. An amount of suboptimality remains that is referred to as the split loss [2] It may be partially counteracted by classified vector quantization (VQ) [3] and by combining split VQ with multistage VQ [4].

DISTORTION MEASURES AND VECTOR PARTITIONING

RENORMALIZATION OF SUBVECTORS

PREDICTIVE SPLIT QUANTIZATION

QUANTIZER TRAINING AND EVALUATION

CONCLUSION