Fixed Codebook Search Using Toeplitz Matrix Approximation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In speech coding apparatus, the processing load increases when a filter with non-causal impulse response cannot be represented by a lower triangular Toeplitz matrix, leading to higher memory and computational requirements.
Innovation Solution
A fixed codebook searching apparatus and method that approximates the transfer function using a Toeplitz-type convolution matrix by truncating row elements from a lower triangular Toeplitz matrix, allowing for efficient convolution processing without significant increases in memory or computational loads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a filter with non-causal impulse response is applied to the excitation pulse, then the signal characteristics and coding quality are improved, but the memory and computational loads increase significantly
Solution Approach 1:
The patent segments the convolution operation into two parts: first computing the correlation between the excitation pulse and the impulse response to obtain correlation coefficients, then using these coefficients to generate the filtered excitation signal. This segmentation transforms the computationally intensive direct convolution into a two-stage process that reduces computational complexity while maintaining the ability to handle non-causal filters
Solution Approach 2:
The patent introduces correlation coefficients as an intermediary variable between the excitation pulse and the filtered excitation signal. Instead of directly convolving the excitation pulse with the non-causal impulse response, the system first computes correlation coefficients that capture the essential relationship, then uses these coefficients to generate the final signal. This intermediary approach significantly reduces the computational burden of handling non-causal filters
2Manufacturing precision
If a filter with non-causal impulse response is applied to the excitation pulse, then the signal characteristics are improved, but the memory requirements increase
Solution Approach 1:
The patent performs preliminary computation of correlation coefficients between the excitation pulse and the impulse response before generating the filtered excitation signal. By pre-computing these correlation coefficients, the system avoids the need to store and process the entire non-causal impulse response during the signal generation phase, thereby reducing memory requirements while preserving the signal characteristics that depend on the non-causal filter
Data Source
AI summary
A fixed codebook searching apparatus, includes a convolution operator, implemented by at least one processor, that convolves an impulse response of a perceptually weighted synthesis filter with an impulse response vector that has values at negative times, to generate a second impulse response vector that has values at negative times. A matrix generator, implemented by at least one processor, generates a Toeplitz-type convolution matrix using the second impulse response vector generated by the convolution operator. A searcher, implemented by at least one processor, performs a codebook search by maximizing a term using the Toeplitz-type convolution matrix.