Audio Encoder Bit Allocation Using MSB-LSB Spectral Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio encoding technologies face challenges in achieving a balanced tradeoff between audio quality, computational complexity, and bitrate, particularly at varying bitrates, due to suboptimal global-gain selection and inefficient handling of high-frequency coefficients.
Innovation Solution
An audio decoder and encoder system that selectively encodes and decodes most significant bits while omitting least significant bits for spectral values, adapting to bit budget constraints through switchable modes and efficient bit allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If high-frequency non-zero coefficients are truncated by setting them to zero to stay within bit budget, then bitrate constraint is satisfied, but codec performance severely degrades at high bitrates
Solution Approach 1:
The patent segments the bitstream into two separate sequences: one for most significant bits (MSBs) and another for least significant bits (LSBs). This segmentation allows independent handling of different bit portions, enabling the decoder to reconstruct truncated coefficients by combining MSBs with available LSBs, thereby maintaining codec performance at high bitrates while satisfying bitrate constraints through selective LSB omission.
Solution Approach 2:
The patent changes the parameter representation by separating coefficients into MSB and LSB components. By encoding MSBs for all non-zero coefficients and selectively encoding LSBs based on bit budget, the system adapts the encoding parameters dynamically. This parameter change enables flexible bitrate control without permanently degrading coefficient representation, as LSBs can be recovered when bandwidth permits.
2Quantity of substance
If global-gain is increased to reduce number of bits consumed by arithmetic encoder, then bitrate is reduced, but distortion introduced by quantization increases resulting in worse perceptual quality
Solution Approach 1:
The patent segments each coefficient into MSB and LSB portions, with MSBs encoded using arithmetic encoding and LSBs encoded separately. This segmentation allows the system to control bit consumption at different precision levels. By encoding MSBs for all coefficients and selectively encoding LSBs based on bit budget, the system reduces total bit consumption while maintaining perceptual quality through preserved MSB information.
Solution Approach 2:
The patent applies different encoding strategies to different parts of the coefficient data. MSBs are encoded with high fidelity using arithmetic encoding for all non-zero coefficients, while LSBs are encoded selectively based on bit budget and perceptual importance. This local quality differentiation ensures that critical high-order information is always preserved while lower-order information is adapted to bitrate constraints.
3Manufacturing precision
If iterative rate-loop approach is used to find optimal global-gain, then audio quality is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary encoding of MSBs for all non-zero coefficients before deciding on LSB encoding. By pre-encoding the most critical bit portion and calculating the bit consumption, the system can determine the remaining bit budget for LSBs without requiring full iterative rate-loop optimization. This preliminary action reduces computational complexity while maintaining the ability to optimize overall audio quality within the bitrate constraint.
Solution Approach 2:
The patent extracts the MSB encoding process from the full iterative rate-loop optimization. By separating MSB encoding (which is always performed) from LSB encoding (which is conditionally performed), the system removes the need for iterative optimization of all bits. This extraction reduces computational complexity by eliminating redundant iterations while preserving the ability to optimize audio quality through selective LSB encoding based on the remaining bit budget.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An audio decoder for providing a decoded audio information on the basis of an encoded audio information is configured to obtain decoded spectral values on the basis of an encoded information representing the spectral values. The audio decoder is configured to jointly decode two or more most significant bits per spectral value on the basis of respective symbol codes for a set of spectral values using an arithmetic decoding, wherein a respective symbol code represents two or more most significant bits per spectral value for one or more spectral values.The audio decoder is configured to decode one or more least significant bits associated with one or more of the spectral values in dependence on how much least significant bit information is available, such that one or more least significant bits associated with one or more of the spectral values are decoded, while no least significant bits are decoded for one or more other spectral values for which one or more most significant bits are decoded and which comprise more bits than the one or more most significant bits. The audio decoder is configured to provide the decoded audio information using the spectral values. Modifications of the audio decoder are possible. An audio encoder is also described.