Audio Encoding Downmix Gain Compensation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Multi-channel audio signals experience sound level loss when downmixed to mono or stereo, leading to degradation in sound quality due to limited bit size and high sound levels, especially when a large number of channels are downmixed.

Innovation Solution

Applying an arbitrary downmix gain to the downmix signal to modify its sound level, using spatial information such as channel level difference, inter-channel coherence, and channel time difference to maintain audio quality during encoding and decoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If a multi-channel audio signal is downmixed to mono or stereo, then the signal can be encoded with limited bit size, but sound level loss occurs and sound quality degrades

Engineering Contradiction:
Improvebit sizeVSAvoidsound quality
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent applies preliminary action by calculating and storing the downmix gain value before the downmixing process. This pre-calculated gain information is then used to compensate for sound level loss during or after downmixing, preventing quality degradation before it occurs rather than correcting it afterward.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter of gain compensation by introducing a downmix gain value that is calculated based on the relationship between original multi-channel signals and downmixed signals. This parameter is applied to adjust the amplitude of downmixed channels, thereby maintaining sound quality while enabling compact encoding.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If downmixing is applied to reduce channel count, then data compression is achieved, but spatial information and sound level are lost

Engineering Contradiction:
Improveencoding efficiencyVSAvoidspatial information
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent implements feedback by calculating the downmix gain based on the actual downmixed signal characteristics and using this calculated gain to adjust the downmixed signal. This closed-loop approach ensures that spatial information is preserved by continuously adapting the gain compensation to the specific signal characteristics.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The downmix gain is calculated in advance based on the relationship between original and downmixed signals, preserving spatial information characteristics before they are lost in the compression process. This preliminary calculation of gain parameters enables efficient encoding while maintaining quality.

Inventive Principle:
Principle #10Preliminary action

3Use of energy by moving object

If high sound levels are present in multi-channel audio, then dynamic range is maintained, but downmixing causes clipping and distortion

Engineering Contradiction:
Improvedynamic rangeVSAvoidclipping and distortion
Core Design Contradiction:
Use of energy by moving objectVSObject-affected harmful factors

Solution Approach 1:

The patent changes the amplitude parameter of downmixed signals by applying the calculated downmix gain, which prevents clipping and distortion while maintaining the dynamic range characteristics of the original multi-channel audio. This parameter adjustment occurs in a controlled manner based on signal relationships.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent converts the potential harm of sound level loss into a benefit by calculating the downmix gain from the relationship between original and downmixed signals. This approach uses the downmixing process itself to generate the compensation information needed to prevent distortion, turning a problematic effect into a useful feature.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentEP1913577B1Apparatus for encoding an audio signal and method thereof
Publication Date: 2021.05.05 LG ELECTRONICS INC
  • EP1913577B1 patent drawingFigure 1
  • EP1913577B1 patent drawingFigure 2(a)~2(c)
  • EP1913577B1 patent drawingFigure 3

AI summary

A method and/or apparatus for encoding and/or decoding an audio signal is disclosed, in which a downmix gain is applied to a downmix signal in an encoding apparatus which, in turn, transmits, to a decoding apparatus, a bitstream containing information as to the applied downmix gain. The decoding apparatus recovers the downmix signal, using the downmix gain information. A method and/or apparatus for encoding and/or decoding an audio signal is also disclosed, in which the encoding apparatus can apply an arbitrary downmix gain (ADG) to the downmix signal, and can transmit a bitstream containing information as to the applied ADG to the decoding apparatus. The decoding apparatus recovers the downmix signal, using the ADG information. A method and/or apparatus for encoding and/or decoding an audio signal is also disclosed, in which the method and/or apparatus can also vary the energy level of a specific channel, and can recover the varied energy level .