Audio Decoder Decorrelation Control via Weighting Factors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Object-based audio systems face challenges in controlling decorrelation at low target bit rates, leading to increased correlation between decoded objects and complex operations depending on rendering, necessitating a simpler and more flexible method for improved reconstruction.
Innovation Solution
The proposed method involves using two weighting factors for approximated and decorrelated audio objects, allowing flexible control of decorrelation, with a single parameter describing the mixture of 'dry' and 'wet' contributions per time/frequency tile, and a reconstruction matrix that is time and frequency variant, enabling reduced unwanted correlation and energy preservation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If decorrelators are introduced in the decoder to improve reconstruction of audio objects, then the correlation between decoded objects is improved, but the device complexity and computational operations increase
Solution Approach 1:
The patent applies parameter changes by introducing weighting factors (w0, w1) that control the amount of decorrelation applied to audio objects. These parameters allow flexible adjustment of the decorrelation strength, enabling the system to optimize between reconstruction quality and computational complexity based on specific requirements. The weighting factors modify the decorrelation process by blending the original and decorrelated signals in controllable proportions.
2Manufacturing precision
If known methods for object-based audio systems are used, then audio objects can be reconstructed, but the methods are sensitive to the number of downmix signals and objects, and may be complex operations depending on rendering
Solution Approach 1:
The patent implements dynamics by making the decorrelation process adaptable through time-varying and frequency-varying weighting factors. These parameters can be adjusted independently for different time/frequency tiles, allowing the system to dynamically optimize reconstruction quality for varying audio content and playback conditions without being constrained by the number of downmix signals or objects.
3Ease of operation
If a single parameter is used to describe the mixture of dry and wet contributions per time/frequency tile, then the control of decorrelation is simplified, but the precision of decorrelation control may be reduced
Solution Approach 1:
The patent applies segmentation by dividing the audio spectrum into multiple time/frequency tiles and applying independent weighting factors to each tile. This segmentation allows precise control of decorrelation at different frequency regions and time intervals while maintaining manageable complexity through the single parameter representation per tile. Each tile can be optimized independently based on local audio characteristics.
Data Source
AI summary
The present disclosure provides methods, devices and computer program products which provide less complex and more flexible control of the introduced decorrelation in an audio coding system. According to the disclosure, this is achieved by calculating and using two weighting factors, one for an approximated audio object and one for a decorrelated audio object, for introduction of decorrelation of audio objects in the audio coding system.