Audio Subband Window Interpolation for Lower Memory and Delay

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Modern digital audio processing systems face challenges in balancing bit rate, computational complexity, memory requirements, quality, and delay, particularly in real-time applications, where compromises often need to be made among these parameters to achieve optimal performance.

Innovation Solution

The use of an interpolation scheme to derive a window function with a smaller number of window coefficients from a larger one, allowing for improved energy distribution and reduced memory usage while maintaining quality and reducing delay, is employed in both analysis and synthesis filterbanks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a larger window function with more coefficients is used, then frequency response accuracy and reconstruction quality are improved, but memory requirements and computational complexity increase

Engineering Contradiction:
Improvefrequency response accuracyVSAvoidmemory requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the window function coefficients into two groups: first group coefficients and second group coefficients. The first group coefficients are used for current frame processing while the second group coefficients are used for next frame processing. This segmentation allows using fewer coefficients per group while maintaining overall frequency response accuracy, thereby reducing memory requirements and computational complexity per frame.

Inventive Principle:
Principle #1Segmentation

2Manufacturing precision

If a larger window function with more coefficients is used, then reconstruction quality is improved, but computational complexity increases

Engineering Contradiction:
Improvereconstruction qualityVSAvoidcomputational complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the window function into two groups of coefficients that are applied alternately to current and next frames. This segmentation reduces the number of coefficients that need to be processed in each individual frame, thereby reducing computational complexity per frame while maintaining overall reconstruction quality through the combined effect of both coefficient groups across frames.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent dynamically switches between using first group coefficients for current frames and second group coefficients for next frames. This dynamic alternation allows the system to maintain high reconstruction quality by utilizing the full set of window function coefficients across frames while reducing the computational burden on any single frame processing operation.

Inventive Principle:
Principle #15Dynamics

3Manufacturing precision

If more window coefficients are used, then audio quality is improved, but delay increases

Engineering Contradiction:
Improveaudio qualityVSAvoiddelay
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent segments the window function coefficients into two groups that can be applied to current and next frames respectively. This segmentation enables the system to achieve high audio quality through the combined use of both coefficient groups while reducing delay by processing each frame with only one group of coefficients, thereby enabling faster processing and reduced buffering requirements.

Inventive Principle:
Principle #1Segmentation

4Reliability

If a full window function is used for each frame, then processing completeness is improved, but memory usage increases

Engineering Contradiction:
Improveprocessing completenessVSAvoidmemory usage
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent divides the window function coefficients into two separate groups: first group coefficients and second group coefficients. Each group is stored separately in memory and applied to alternating frames (current frame and next frame respectively). This segmentation reduces the memory footprint for each frame's window coefficients while ensuring processing completeness is maintained through the systematic alternation between the two coefficient groups across consecutive frames.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP1994530B1Apparatus and method for generating audio subband values and apparatus and method for generating time-domain audio samples
Publication Date: 2009.07.01 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP1994530B1 patent drawingFigure 1
  • EP1994530B1 patent drawingFigure 2A
  • EP1994530B1 patent drawingFigure 2B

AI summary

An embodiment of an apparatus (100) for generating audio subband values in audio subband channels comprises an analysis windower (110) for windowing a frame (120) of time-domain audio input samples being in a time sequence extending from an early sample to a later sample using an analysis window function (190) comprising a sequence of window coefficients to obtain windowed samples. The analysis window function comprises a first number of window coefficients derived from a larger window function comprising a sequence of a larger second number of window coefficients, wherein the window coefficients of the window function are derived by an interpolation of window coefficients of the larger window function. The apparatus (100) further comprises a calculator (170) for calculating the audio subband values using the windowed samples.