VARIABLE-DIMENSION VECTOR QUANTIZATION

In many signal compression applications, the evolution of the signal o ver time can be represented by a sequence of random vectors with varyi ng dimensionality. Frequently, the generation of such variable-dimensi on vectors can be modeled as a random sampling of another signal vecto r with a large but fixed dimension. Efficient quantization of these va riable-dimension vectors is a challenging task and a critical issue in speech coding algorithms based on harmonic spectral modeling. We intr oduce a simple and effective formulation of the problem and present a novel technique, called variable-dimension vector quantization (VDVQ), where the input variable-dimension vector is directly quantized with a single universal codebook. The application of VDVQ to low bit-rate s peech coding demonstrates significant gain in subjective quality as we ll as in rate-distortion performance over prior indirect methods.