SBR Audio Decoding with Low-Band Smoothing for HF Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional audio signal decoding methods using SBR (Spectral Band Replication) often result in auditory degradation due to energy depressions in low-frequency range signals, leading to inaccuracies in high-frequency range signal reproduction and reduced audio quality.
Innovation Solution
A signal processing method that includes flattening (smoothing) the low-frequency range signal to remove energy depressions, followed by frequency shifting and gain adjustment to generate high-frequency range signals, ensuring the average energy of the resulting signal matches the high-frequency range scalefactor band energy, thereby maintaining the original frequency shape.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SBR is used to generate high-range signals from low-range signals, then audio quality is improved, but energy depressions in low-range signals cause auditory degradation and inaccuracies in high-frequency reproduction
Solution Approach 1:
The patent applies smoothing processing to the low-range signal before frequency shifting to generate high-range signals. This preliminary action removes energy depressions and irregularities in the low-range signal, ensuring that the subsequent frequency-shifted high-range signals are free from artifacts and accurately represent the intended audio spectrum, thereby resolving the contradiction between reproduction accuracy and quality consistency
Solution Approach 2:
The patent introduces smoothing processing as an intermediary step between the low-range signal and the high-range signal generation. This intermediary process filters out energy depressions and irregularities, creating a cleaner base signal for frequency shifting, which improves both the accuracy of high-frequency reproduction and the consistency of overall audio quality
2Productivity
If gain limiting or interpolation is applied during high-range signal generation, then processing efficiency is improved, but energy depressions cause auditory degradation
Solution Approach 1:
The patent applies smoothing processing before gain limiting or interpolation operations. By removing energy depressions in advance, the subsequent processing steps can be performed efficiently without introducing auditory degradation, thus maintaining both processing efficiency and audio quality
Solution Approach 2:
Smoothing processing serves as an intermediary that prepares the low-range signal by eliminating energy depressions before efficiency-optimizing operations like gain limiting or interpolation are applied. This ensures that these efficient processing steps do not amplify or propagate quality degradation
3Adaptability or versatility
If frequency shifting is applied to generate high-range signals, then band expansion is achieved, but energy depressions in low-range signals are reproduced in high-range signals
Solution Approach 1:
The patent applies smoothing processing to remove energy depressions from the low-range signal before frequency shifting is applied. This preliminary action ensures that when the signal is frequency-shifted to generate high-range components, the irregularities and depressions are not carried over, maintaining accurate frequency spectrum representation while achieving broad band coverage
Solution Approach 2:
Smoothing processing acts as an intermediary between the low-range signal and frequency shifting operations. It filters out energy depressions that would otherwise be reproduced in the high-range signals, allowing band expansion to occur without compromising frequency spectrum accuracy
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A method, system, and computer program product for processing an encoded audio signal is described. In one exemplary embodiment, the system receives an encoded low-frequency range signal and encoded energy information used to frequency shift the encoded low-frequency range signal. The low-frequency range signal is decoded and an energy depression of the decoded signal is smoothed. The smoothed low-frequency range signal is frequency shifted to generate a high-frequency range signal. The low-frequency range signal and high-frequency range signal are then combined and outputted.