Reversible Integer-Integer Overlapped Transform for Audio Pre-Echo Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio compression systems face challenges with pre-echo effects due to sudden changes in audio signals, particularly when using transform coding with fixed frame sizes, which lead to quantization noise spreading across the entire block, affecting auditory quality.
Innovation Solution
Implementing a reversible integer-integer overlapped transform with variable subframe sizes, decomposing the Modulated Lapped Transform (MLT) into a combination of overlapping and block transforms, allowing for adaptive window sizes to reduce pre-echo and improve coding efficiency and quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If fixed frame sizes are used in transform coding, then coding efficiency is improved, but pre-echo effects occur due to quantization noise spreading across the entire block
Solution Approach 1:
The patent divides the audio signal into variable-sized subframes rather than using fixed frame sizes. This segmentation allows different portions of the audio to be processed with appropriate resolutions, preventing quantization noise from spreading across entire blocks and causing pre-echo effects while maintaining coding efficiency.
Solution Approach 2:
The patent implements variable subframe sizes that adapt to the audio signal characteristics. By dynamically adjusting frame sizes based on the audio content, the system optimizes the balance between coding efficiency and preventing pre-echo effects, allowing larger frames for stable regions and smaller frames for transient passages.
2Object-affected harmful factors
If overlapping transforms are used to reduce blocking artifacts, then auditory quality is improved, but device complexity increases due to variable window frame sizes
Solution Approach 1:
The patent segments the transform process into manageable components with variable window sizes that overlap to reduce blocking artifacts. By dividing the audio into overlapping subframes with appropriate windowing, the system reduces quantization discontinuities at block boundaries while maintaining implementation feasibility through systematic segmentation.
Solution Approach 2:
The patent changes the window size parameter dynamically to optimize the balance between reducing blocking artifacts and managing computational complexity. By adjusting window dimensions based on signal characteristics, the system achieves effective artifact reduction without excessive computational burden.
3Productivity
If larger windows are used in transform coding, then coding efficiency is improved, but time resolution deteriorates
Solution Approach 1:
The patent dynamically adjusts window sizes based on the audio signal characteristics and processing requirements. For stationary regions, larger windows provide better coding efficiency, while for transient passages, smaller windows maintain time resolution. This dynamic adaptation resolves the contradiction between efficiency and resolution.
Solution Approach 2:
The patent applies different window sizes to different portions of the audio signal based on their local characteristics. By tailoring the window dimensions to the specific requirements of each segment, the system optimizes both coding efficiency and time resolution where needed, rather than using a uniform approach throughout.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An audio encoder/decoder uses a combination of an overlap windowing transform and block transform that have reversible implementations to provide a reversible, integer-integer form of a lapped transform. The reversible lapped transform permits both lossy and lossless transform domain coding of an audio signal having variable subframe sizes.