Spectral Postprocessing for Scalable Lossless Audio Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio encoding and decoding technologies, such as MP3 and AAC, face challenges in achieving lossless compression due to the incompatibility of their filter banks with integer transformations, leading to inefficient scalability and increased computational overhead, especially in mobile devices.
Innovation Solution
A postprocessing method that uses weighted additions of spectral values from one transformation algorithm to approximate results from a different transformation algorithm, such as integer MDCT, allowing for the formation of an efficient extension layer without the need for additional decoding stages or bit-accurate decoders.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If MP3 or AAC encoding standards are used for audio compression, then audio signal compression is achieved, but lossless compression and scalability are limited due to filter bank incompatibility with integer transformations
Solution Approach 1:
The patent introduces an intermediary postprocessing stage that converts spectral values from the first transformation algorithm (MP3 hybrid filter bank or AAC MDCT) into a form compatible with the second transformation algorithm (integer MDCT). This intermediary conversion layer enables lossless extension by bridging the incompatibility between different transformation algorithms without requiring complete re-encoding or bit-accurate decoders, thus reducing decoder complexity while achieving lossless compression.
2Measurement precision
If scalable encoding with base layer and extension layer is implemented, then audio quality is improved, but computational overhead increases especially in mobile devices
Solution Approach 1:
The patent performs preliminary postprocessing of spectral values from the base layer encoding to transform them into a format compatible with integer MDCT before the extension layer processing. This preliminary conversion prepares the data in advance, avoiding the need for complex real-time computations during decoding in mobile devices, thus reducing computational energy while maintaining high audio quality through scalable encoding.
3Adaptability or versatility
If different transformation algorithms are used for base layer and extension layer, then scalability is achieved, but spectral value incompatibility prevents efficient encoding
Solution Approach 1:
The patent changes the parameters of spectral values from the first transformation algorithm through weighted addition operations to match the parameters expected by the second transformation algorithm. By adjusting spectral value parameters (amplitude, phase, frequency distribution) through the postprocessing conversion, the patent enables efficient encoding with different transformation algorithms in the base and extension layers, achieving both scalability and encoding efficiency.
Data Source
AI summary
For postprocessing spectral values which are based on a first transformation algorithm for converting the audio signal into a spectral representation, first a sequence of blocks of the spectral values representing a sequence of blocks of samples of the audio signal are provided. Hereupon, a weighted addition of spectral values of the sequence of blocks of spectral values is performed in order to obtain a sequence of blocks of postprocessed spectral values, wherein the combination is performed such that for calculating a postprocessed spectral value for a frequency band and a time duration a spectral value of the sequence of blocks for the frequency band and the time duration and a spectral value for another frequency band or another time duration are used, wherein the combination is further performed such that such weighting factors are used that the postprocessed spectral values are an approximation to the spectral values as they are obtained by converting the audio signal into a spectral representation using a second transformation algorithm which is different from the first transformation algorithm. The postprocessed spectral values are in particular used for a difference formation within a scalable encoder or for an addition within a scalable decoder, respectively.