Multi-Channel Audio Encoding With Selective Subband Reverberation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-channel audio encoding methods require a large number of bits to encode reverberation gain parameters for each subband, leading to reduced auditory quality and increased bit overhead, especially when the coherence between channel signals is low, resulting in distorted restored signals.
Innovation Solution
A method that determines a target reverberation gain parameter based on energy differences and coherence between channel signals, allowing only specific subbands to be encoded, reducing bit usage and improving encoding efficiency by allocating bits more effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If reverberation gain parameters corresponding to all subbands are encoded, then the auditory quality is improved, but the bit overhead increases significantly
Solution Approach 1:
The patent divides the frequency spectrum into multiple subbands and selectively encodes reverberation gain parameters only for specific subbands (e.g., low-frequency subbands) rather than all subbands. This segmentation approach reduces the total number of parameters to be encoded, thereby reducing bit overhead while maintaining auditory quality in the most perceptually important frequency ranges.
Solution Approach 2:
The patent applies different encoding strategies to different subbands based on their perceptual importance. Specifically, reverberation gain parameters are encoded for low-frequency subbands where human hearing is most sensitive, while high-frequency subbands use default or simplified parameters. This local quality approach ensures high auditory quality where it matters most while minimizing overall bit consumption.
2Measurement precision
If reverberation processing is performed on each subband, then the spatial accuracy is improved, but the encoding efficiency deteriorates
Solution Approach 1:
The patent extracts and processes only the most critical subbands (particularly low-frequency subbands) for reverberation gain parameter encoding, while omitting processing for less important high-frequency subbands. This selective extraction maintains spatial accuracy for the most perceptually relevant components while significantly improving encoding efficiency by reducing the number of processing operations required.
Solution Approach 2:
The patent applies partial action by performing reverberation processing on only a subset of subbands rather than all subbands. Specifically, it focuses processing resources on low-frequency subbands where spatial accuracy is most critical for human perception, while using default parameters for high-frequency subbands, thus achieving acceptable spatial accuracy with improved encoding efficiency.
3Measurement precision
If more bits are allocated to encode all reverberation gain parameters, then the signal fidelity is improved, but the overall encoding performance deteriorates due to inefficient bit allocation
Solution Approach 1:
The patent changes the parameter encoding strategy by selectively applying different levels of precision to different subbands. Instead of uniformly encoding all reverberation gain parameters with the same bit allocation, it uses higher precision for low-frequency subbands and lower or default precision for high-frequency subbands. This parameter change optimizes the trade-off between signal fidelity and overall encoding performance by allocating bits according to perceptual importance.
Data Source
AI summary
A multi-channel signal decoding method includes obtaining a downmixed signal of a first channel signal and a second channel signal in a multi-channel signal, obtaining parameter indication information, obtaining a reverberation gain parameter corresponding to each of the part of subbands according to the parameter indication information; and obtaining a reconstructed first channel signal and a reconstructed second channel signal according to the downmixed signal and the reverberation gain parameter.