Two-Channel Audio Decoding With Intelligent Gap Filling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio codecs face limitations in encoding and decoding two-channel representations, particularly in handling different correlation situations between source and target ranges, leading to artifacts such as spatial segregation and increased computational complexity due to the need for transformation into new domains for bandwidth extension.
Innovation Solution
The proposed solution involves calculating parametric data for reconstruction bands to identify suitable two-channel representations and using Intelligent Gap Filling (IGF) to regenerate spectral portions, allowing for efficient encoding and decoding in the same spectral domain without the need for downsampling or upsampling, and employing frequency tile filling to fill spectral gaps while preserving tonal components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If bandwidth extension techniques are used to encode high-frequency content from low-frequency regions, then coding efficiency is improved and bitrate is reduced, but artifacts such as spatial segregation occur and audio quality deteriorates
Solution Approach 1:
The patent applies local quality by differentiating between correlated and uncorrelated spectral portions. For correlated portions, spectral copying is performed; for uncorrelated portions, noise-like filling is applied. This localized approach ensures that artifacts only appear where spectrally appropriate, preserving audio quality in regions where correlation exists while maintaining coding efficiency where it doesn't.
Solution Approach 2:
The patent dynamically adapts the bandwidth extension strategy based on the correlation characteristics of each spectral portion. The system switches between spectral copying and noise-like filling depending on the measured correlation, allowing flexible adaptation to different signal conditions and preventing spatial segregation artifacts in correlated regions.
2Productivity
If spectral copying is performed for bandwidth extension, then coding efficiency is improved, but artifacts occur in uncorrelated spectral portions
Solution Approach 1:
The patent applies local quality by differentiating between correlated and uncorrelated spectral portions. For correlated portions, spectral copying is performed; for uncorrelated portions, noise-like filling is applied. This localized approach ensures that artifacts only appear where spectrally appropriate, preserving audio quality in regions where correlation exists while maintaining coding efficiency where it doesn't.
Solution Approach 2:
The patent uses feedback by calculating correlation measures between low-frequency and high-frequency spectral portions and using this information to determine the appropriate filling strategy. This feedback mechanism allows the system to adapt its bandwidth extension approach based on actual signal characteristics, preventing artifacts in uncorrelated regions.
3Reliability
If transformation into new domains is performed for bandwidth extension, then high-frequency content can be reconstructed, but computational complexity increases
Solution Approach 1:
The patent merges the bandwidth extension process with the existing MDCT coding framework, performing all operations in the same spectral domain. This eliminates the need for separate transform domains and reduces computational complexity while maintaining reliable high-frequency reconstruction through spectral copying and noise-like filling.
Data Source
Figure 1A~1B
Figure 2A
Figure 2B
AI summary
An apparatus for generating a decoded two-channel signal, comprises: an audio processor (802) for decoding an encoded two-channel signal to obtain a first set of first spectral portions; a parametric decoder (804) for providing parametric data for a second set of second spectral portions and a two-channel identification identifying either a first or a second different two-channel representation for the second spectral portions; and a frequency regenerator (806) for regenerating a second spectral portion depending on a first spectral portion of the first set of first spectral portions, the parametric data for the second portion and the two-channel identification for the second portion.