Audio Coding Using Interpolated Power Limits to Cut Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio coding methods, such as MPEG-1 Layer III, face challenges in achieving low delay times while maintaining high compression efficiency and minimizing audible artifacts, particularly due to the need to transfer masking thresholds and filter coefficients, which can lead to increased bit rate and audible interferences.
Innovation Solution
The proposed solution involves applying a psycho-acoustic model to multiple blocks of audio values to calculate parameterizations and noise power limits, interpolating these values to derive an interpolated parameterization and noise power limit, and using these to filter and scale audio signals, thereby reducing audible artifacts and optimizing bit rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If masking thresholds and filter coefficients are transferred to achieve high compression efficiency, then compression efficiency is improved, but bit rate increases and audible interferences occur
Solution Approach 1:
The patent extracts only the essential parameters (noise power limits and interpolated parameterizations) from the complete psycho-acoustic model data, transmitting only what is necessary for reconstruction rather than the full masking thresholds and filter coefficients
Solution Approach 2:
The patent transforms the parameter representation by calculating noise power limits and interpolated parameterizations as alternative representations of the psycho-acoustic model, which require fewer bits to transmit while maintaining compression efficiency
2Productivity
If masking thresholds and filter coefficients are transferred to achieve high compression efficiency, then compression efficiency is improved, but audible interferences increase
Solution Approach 1:
The patent extracts only the essential parameters (noise power limits and interpolated parameterizations) from the complete psycho-acoustic model data, transmitting only what is necessary for reconstruction rather than the full masking thresholds and filter coefficients
Solution Approach 2:
The patent introduces interpolated parameterizations as an intermediary representation that bridges the transmitter and receiver, allowing reconstruction of the psycho-acoustic model without directly transmitting the original masking thresholds and filter coefficients that cause audible interferences
3Loss of time
If short delay time is achieved by using small block sizes, then delay time is reduced, but compression efficiency decreases
Solution Approach 1:
The patent changes the parameter representation to noise power limits and interpolated parameterizations, which can be effectively calculated and transmitted for small block sizes, enabling short delay times while maintaining compression efficiency through the interpolated reconstruction approach
Data Source
AI summary
The central idea of the present invention is that the prior procedure, namely interpolation relative to the filter coefficients and the amplification value, for obtaining interpolated values for the intermediate audio values starting from the nodes has to be dismissed. Coding containing less audible artifacts can be obtained by not interpolating the amplification value, but rather taking the power limit derived from the masking threshold, for each node, i.e. for each parameterization to be transferred, and then performing the interpolation between these power limits of neighboring nodes, such as, for example, a linear interpolation. On both the coder and the decoder side, an amplification value can then be calculated from the intermediate power limit determined such that the quantizing noise caused by quantization, which has a constant frequency before post-filtering on the decoder side, is below the power limit or corresponds thereto after post-filtering.