Optimized Scale Factor for Audio Frequency Band Extension
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing AMR-WB codec's band extension technique faces issues with subframe gain estimation, leading to suboptimal high-band excitation levels, energy overestimation, and desynchronization between low and high bands, resulting in degraded audio quality, especially at higher bit rates.
Innovation Solution
A method to determine an optimized scale factor for the excitation signal or filter in the frequency band extension process, using a lower-order additional filter to equalize gains between frequency bands, computed based on the coefficients of the decoded linear prediction filter, to avoid energy overestimation and improve prediction filter equalization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the conventional band extension technique is used with subframe gain estimation, then the high-band excitation can be generated, but energy overestimation and desynchronization between low and high bands occur, degrading audio quality
Solution Approach 1:
The patent changes the parameter used for gain estimation from subframe-based estimation to frame-based estimation with a global scale factor. This parameter change eliminates the desynchronization issue between low and high bands while reducing energy overestimation, thereby improving audio quality without sacrificing gain estimation accuracy
Solution Approach 2:
The patent extracts the gain estimation process from the subframe level and moves it to the frame level. By separating the global scale factor determination from the subframe processing, the patent avoids the synchronization problems that arise when gain is estimated at the subframe level, while maintaining accurate energy representation
2Productivity
If the band extension is performed with subframe-based gain estimation, then the processing can be done in real-time, but desynchronization between low and high bands occurs
Solution Approach 1:
The patent merges the gain estimation process from the subframe level to the frame level, creating a unified global scale factor that applies to the entire frame. This merging eliminates the synchronization instability between low and high bands while maintaining real-time processing capability, as the frame-based approach reduces the number of gain calculations needed
3Device complexity
If the conventional scale factor is used for excitation signal, then the implementation is simple, but energy overestimation occurs leading to degraded audio quality
Solution Approach 1:
The patent introduces a new parameter - the global scale factor - that replaces the conventional subframe-based gain estimation. This parameter change maintains implementation simplicity while eliminating energy overestimation, thereby improving audio quality without significantly increasing device complexity
Data Source
AI summary
A method and device are provided for determining an optimized scale factor to be applied to an excitation signal or a filter during a process for frequency band extension of an audio frequency signal. The band extension process includes decoding or extracting, in a first frequency band, an excitation signal and parameters of the first frequency band including coefficients of a linear prediction filter, generating an excitation signal extending over at least one second frequency band, filtering using a linear prediction filter for the second frequency band. The determination method includes determining an additional linear prediction filter, of a lower order than that of the linear prediction filter of the first frequency band, the coefficients of the additional filter being obtained from the parameters decoded or extracted from the first frequency and calculating the optimized scale factor as a function of at least the coefficients of the additional filter.