Adaptive Time Resolution Audio Coding for Pre-Echo Suppression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio coding techniques, such as transform coders, face challenges in mitigating pre-echo artifacts, particularly in handling transients, due to increased complexity, bit-rate overhead, and delayed processing, which are not effectively addressed by current methodologies like bit reservoir techniques, gain modification, temporal noise shaping, and window switching.

Innovation Solution

The method involves using time-domain aliased frames for segmentation and spectral analysis, allowing for adaptive time segmentation and spectral analysis based on sub-frames, which enables instantaneous switching to higher time resolution and efficient bit-rate encoding, thereby mitigating pre-echo effects without the need for additional delay or increased complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If window switching is used to mitigate pre-echo artifacts, then pre-echo suppression is improved, but device complexity and processing delay increase

Engineering Contradiction:
Improvepre-echo artifactVSAvoidcodec complexity
Core Design Contradiction:
Object-affected harmful factorsVSDevice complexity

Solution Approach 1:

The patent applies dynamics by making the transform block size adaptive rather than fixed. The codec dynamically adjusts the block size between long and short blocks based on signal characteristics (transient detection), allowing optimal time resolution for pre-echo suppression while managing complexity through controlled variability in processing parameters

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent segments the audio signal into different transform blocks (long and short blocks) with different time resolutions. By dividing the processing into segments with appropriate block sizes, the system can apply fine time resolution only where needed (transient regions) rather than uniformly across all signal portions, reducing overall complexity

Inventive Principle:
Principle #1Segmentation

2Object-affected harmful factors

If transform block size is reduced to enable temporal pre-masking, then pre-echo mitigation is improved, but coding gain decreases

Engineering Contradiction:
Improvepre-echo distortionVSAvoidcoding gain
Core Design Contradiction:
Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The patent applies local quality by using different transform block sizes for different portions of the signal. Short blocks with fine time resolution are used locally in transient regions where pre-echo occurs, while long blocks with coarse time resolution are used in stationary regions to maintain coding gain. This localized adaptation optimizes both pre-echo suppression and compression efficiency

Inventive Principle:
Principle #3Local quality

3Object-affected harmful factors

If bit reservoir technique is used to accommodate transient frames, then transient handling is improved, but transmission delay increases

Engineering Contradiction:
Improvetransient distortionVSAvoidtransmission delay
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

The patent uses dynamics by adaptively adjusting transform block size in real-time based on transient detection, rather than using bit reservoir buffering. This allows the system to handle transients immediately with appropriate fine time resolution, avoiding the delay inherent in buffering approaches while still accommodating transient frames effectively

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3550564B1Low-complexity spectral analysis/synthesis using selectable time resolution
Publication Date: 2020.07.22 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3550564B1 patent drawingFigure 1
  • EP3550564B1 patent drawingFigure 2A~2B
  • EP3550564B1 patent drawingFigure 3

AI summary

The signal processing is based on the concept of using a time-domain aliased (12, TDA) frame as a basis for time segmentation (14) and spectral analysis (16), performing segmentation in time based on the time-domain aliased frame and performing spectral analysis based on the resulting time segments. The time resolution of the overall "segmented" time-to-frequency transform can thus be changed by simply adapting the time segmentation to obtain a suitable number of time segments based on which spectral analysis is applied. The overall set of spectral coefficients, obtained for all the segments, provides a selectable time-frequency tiling of the original signal frame.