Speech LSF quantization with rate independent complexity, bit scalability and learning

A computationally efficient, high quality, vector quantization scheme based on a parametric probability density function (PDF) is proposed. In this scheme, speech line spectral frequencies (LSF) are modeled as i.i.d realizations of a multivariate Gaussian mixture density. The mixture model parameters are efficiently estimated using the expectation maximization (EM) algorithm. An efficient quantization scheme using transform coding and bit allocation techniques which allows for easy and computationally efficient mapping from observation to quantized value is developed for both fixed rate and variable rate systems. An attractive feature of this method is that source encoding using the resultant codebook involves very few searches and its computational complexity is minimal and independent of the rate of the system. Furthermore, the proposed scheme is bit scalable and can switch between memoryless and quantizer with memory seamlessly. The performance of the memoryless quantizer is 2-3 bits better than conventional quantization schemes