Audio Object Reconstruction Energy Level Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio object encoding methods, such as Joint Object Coding, often result in level errors like level loss or build-up during reconstruction, leading to inconsistencies in audio playback, particularly when rendered over different speaker configurations like 7.1.4 systems.

Innovation Solution

A method that modifies object reconstruction information by analyzing the difference between original and processed rendered presentations, adjusting reconstruction parameters to align the energy levels of reconstructed objects with those of the original objects, and includes multiple rendering configuration-specific modification gains in the bitstream to mitigate level errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If JOC parameters are adjusted to match object energies, then energy level accuracy is improved, but covariance matrix matching deteriorates

Engineering Contradiction:
Improveenergy level accuracyVSAvoidcovariance matrix matching
Core Design Contradiction:
Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent modifies the JOC parameters by applying energy matching gains derived from the difference between original and reconstructed object energies. This parameter adjustment ensures that the reconstructed objects have energies matching the original objects, resolving the energy level accuracy issue while accepting the trade-off in covariance matrix matching

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent pre-calculates energy matching gains during the encoding phase and stores them in the bitstream. These pre-computed gains cushion against potential energy mismatches during decoding and rendering, ensuring consistent energy levels across different playback configurations before the actual rendering occurs

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

2Productivity

If object-based audio is transmitted efficiently using downmix signals, then transmission efficiency is improved, but reconstruction accuracy deteriorates

Engineering Contradiction:
Improvetransmission efficiencyVSAvoidreconstruction accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent computes energy matching gains by comparing the energies of original objects with reconstructed objects from downmix signals. This feedback mechanism identifies reconstruction errors and compensates for them by adjusting JOC parameters, thereby improving reconstruction accuracy while maintaining the efficiency benefits of downmix-based transmission

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces energy matching gains as an intermediary parameter that bridges the gap between downmix-based transmission and accurate object reconstruction. These gains are transmitted alongside the downmix signals and JOC parameters, serving as a correction factor that improves reconstruction accuracy without compromising transmission efficiency

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If JOC parameters are time- and frequency-varying for each T/F tile, then approximation quality is improved, but data transmission volume increases

Engineering Contradiction:
Improveapproximation qualityVSAvoiddata transmission volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies energy matching only to the diagonal elements of the covariance matrix (object energies) rather than attempting to match the complete covariance matrix. This partial action approach achieves perceptually important energy level accuracy while avoiding the excessive computational and data requirements of complete covariance matrix matching

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4298629B1Audio object processing
Publication Date: 2024.11.06 DOLBY INTERNATIONAL AB
  • EP4298629B1 patent drawingFigure 1
  • EP4298629B1 patent drawingFigure 2A
  • EP4298629B1 patent drawingFigure 2B

AI summary

A method for modifying object reconstruction information, comprising obtaining a set of N spatial audio objects, each spatial audio object including an audio signal and spatial metadata, obtaining an audio presentation representing the N spatial audio objects, obtaining object reconstruction information configured to reconstruct the N spatial audio objects from the audio presentation, applying the reconstruction information to the audio presentation to form a set of N reconstructed spatial audio objects, using a first rendering configuration, rendering the N spatial audio objects to obtain a first rendered presentation, and rendering the N reconstructed spatial audio objects to obtain a second rendered presentation, and modifying the reconstruction information based on a difference between the first rendered presentation and the second rendered presentation, thereby forming modified reconstruction information.