Object-Based Audio Downmix Control Without Sound Distortion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio signal processing methods struggle to control the gain and panning of object-based audio signals without distortion, especially when adjusting vocal or background music levels, and fail to maintain sound quality during large-range adjustments.

Innovation Solution

A method and apparatus that separate downmix information into independent and background objects using enhanced object information, allowing for flexible control of gain and panning, and include a module to generate multi-channel information for adjusting object signals, preventing distortion by subtracting specific signals from the downmix in the time or frequency domain.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If gain adjustment is applied to object-based audio signals, then control flexibility is improved, but sound quality distortion occurs

Engineering Contradiction:
Improvecontrol flexibilityVSAvoidsound quality
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The audio signal is segmented into multiple independent objects (vocal, background music, effects) that can be individually controlled. Each object is processed separately through its own gain adjustment path, allowing independent volume control without affecting other elements. This segmentation enables flexible gain adjustment while maintaining overall sound quality by preventing distortion from cumulative adjustments.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A mixing unit acts as an intermediary between individual object gain adjustments and the final output. The mixing unit receives adjusted objects and combines them in a controlled manner, preventing distortion by managing the interaction between multiple adjusted signals. This intermediary ensures that gain adjustments on individual objects do not result in overall sound quality degradation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If large-range gain adjustment is applied to vocal or background music, then control range is improved, but distortion in sound quality is generated

Engineering Contradiction:
Improvecontrol rangeVSAvoidsound quality
Core Design Contradiction:
Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The system dynamically adjusts gain for each object based on its individual characteristics and the current mixing context. Rather than applying fixed gain adjustments, the system adapts the gain values in real-time to maintain optimal sound quality across the full control range. This dynamic approach allows large-range adjustment while preventing distortion through continuous optimization.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes multiple parameters simultaneously (gain, panning, compression settings) for each object to maintain sound quality during large-range adjustments. When vocal gain is increased significantly, complementary parameter changes are applied to prevent distortion. This multi-parameter adjustment strategy enables extended control range while preserving audio fidelity.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If multiple independent objects are processed individually, then control precision is improved, but system complexity increases

Engineering Contradiction:
Improvecontrol precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

Multiple independently processed objects are merged in the mixing unit into a unified output signal. This merging process consolidates the complexity of individual object processing into a single integration point, maintaining control precision for each object while managing system complexity through centralized mixing. The mixing unit combines all adjusted objects without requiring separate processing paths for each combination scenario.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS9373333B2Method and apparatus for processing an audio signal
Publication Date: 2016.06.21 LG ELECTRONICS INC
  • US9373333B2 patent drawing
  • US9373333B2 patent drawing
  • US9373333B2 patent drawing

AI summary

A method for decoding an audio signal, receiving a downmix signal having at least one independent object and a background object downmixed therein receiving object information and enhanced object information, wherein the object information includes at least one of level information and correlation information between the independent object and the background object, wherein the enhanced object information includes a residual signal extracting the at least one independent object and the background object from the downmix signal using the object information and the enhanced object information receiving mix information from a user, the mix information being usable to control gain or panning of the independent object or the background object generating downmix processing information using at least one of the object information and enhanced object information processing at least one independent object and the background object using at least one of the downmix processing information and the mix information.