Line spectrum-to-parameter dimensional reduction quantizing method based on conditional Gaussian mixture model

Abstract

The invention provides a line spectrum-to-parameter dimensional reduction quantizing method based on a conditional Gaussian mixture model. Specifically, the method comprises the following steps of: firstly framing sampled voice signals, extracting LSP (Linear Spectrum Pair) characteristic parameters of the voice signals to carry out characteristic parameter dimension reduction; then dividing a characteristic parameter sequence to obtain subvector; combining a subvector parameter sequence in pairs, and establishing a union sequence; training a conditional Gaussian mixture model by utilizing the union sequence to obtain the parameters of the conditional Gaussian mixture model; calculating the conditional probability density by utilizing parameters of mean value vector, covariance matrix and the like of the conditional Gaussian mixture model, wherein the number is equal to that of a Gaussian component; then grouping the data, and including the current frame data into a group distributed by the Gaussian component with maximum conditional probability density; training a code book to the grouped data by using an LBG (Linde, Buzo and Gray) algorithm, thus finally obtaining the code book, namely the vector quantizing result of the voice signal. The line spectrum-to-parameter dimensional reduction quantizing method based on the conditional Gaussian mixture model can be used for promoting the quantizing property, and is simple to train, and low in calculation complexity.

Description

technical field [0001] The invention relates to a parameter quantization method, in particular to a parameter dimension reduction quantization method based on a conditional Gaussian mixture model. Background technique [0002] The LSP (Line Spectrum Pair) parameter is an important parameter in speech coding, which plays an absolute role in the speech quality after decoding, so the quantization of this parameter is very important. [0003] The quantization of LSP parameters has been a hot issue since the 1970s. The main direction of the research is the improvement of the structure of the vector quantizer. The initial quantizer is a split vector quantizer, which splits the LSP parameters into several sub-vectors with smaller dimensions, and then trains the codebook and quantizes them separately. This method was groundbreaking at the time, and it greatly reduced the computational complexity while retaining the advantages of vector quantization. A direct development of the spl...

AI Technical Summary

Problems solved by technology

The lattice quantizer is also used to quantize the LSP parameters, but because the lattice quantizer has a better effect on the Gaussian distribution, the distribution of the LSP parameters is not easy to determine

Images