Presentation Loudness Metadata for Consistent Audio Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio encoding and decoding technologies struggle to maintain accurate loudness consistency across different content substreams and presentations, leading to potential inaccuracies that exceed industry tolerances, such as 0.1 dB, which is crucial for maintaining consistent audio levels across various audio content configurations.

Innovation Solution

The proposed solution involves providing loudness data for each presentation, using mixing coefficients and dynamic range compression (DRC) data within the encoding process, allowing decoders to accurately control playback gain and ensure consistent loudness levels by combining substreams based on psychoacoustic modeling, even when user preferences or content changes occur.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If loudness data is provided for each content substream individually, then the decoder can process each substream separately, but the overall loudness accuracy deteriorates because adding individual loudness values does not equal the loudness of the combined signal

Engineering Contradiction:
Improvedecoder processing flexibilityVSAvoidloudness measurement accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent segments the loudness information by providing separate loudness data for each content substream (dialog, music, effects) while also providing an overall presentation loudness value. This allows the decoder to selectively process substreams based on user preferences while maintaining accurate overall loudness control through the presentation-level loudness data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary approach by providing both individual substream loudness data and overall presentation loudness data. The decoder uses the overall presentation loudness as a reference to maintain accurate loudness levels when combining substreams, while individual substream data allows for flexible mixing and selection without compromising overall loudness accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple content substreams are mixed to create different presentations, then content flexibility and adaptability improve, but maintaining consistent loudness across all presentations becomes more difficult

Engineering Contradiction:
Improvepresentation configuration flexibilityVSAvoidloudness consistency
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies universality by providing a standardized loudness data structure that works across all presentations and substream combinations. Each presentation contains loudness data for its constituent substreams plus an overall presentation loudness value, allowing the same decoding approach to maintain loudness consistency regardless of which substreams are active or how they are mixed.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent implements feedback by providing measured loudness values for each presentation configuration in the broadcast signal. The decoder uses these feedback loudness values to adjust its mixing and gain control, ensuring that regardless of which substreams are combined or how mixing coefficients are applied, the output loudness remains consistent with the reference level.

Inventive Principle:
Principle #23Feedback

3Ease of operation

If loudness control is implemented at the decoder, then user flexibility improves, but the complexity of the decoding process increases

Engineering Contradiction:
Improveuser control flexibilityVSAvoiddecoder processing complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-calculating and embedding loudness data for each content substream and each presentation configuration in the broadcast signal before transmission. This allows the decoder to simply retrieve and apply these pre-computed values rather than performing complex loudness measurements and calculations in real-time, reducing decoder complexity while maintaining user flexibility.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements self-service by providing the decoder with all necessary loudness information (individual substream loudness values and overall presentation loudness values) needed to perform accurate loudness control independently. The decoder uses this self-provided information to automatically adjust mixing coefficients and gain levels without requiring external calibration or complex processing algorithms.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4583103B1Transmission-agnostic presentation-based program loudness
Publication Date: 2026.02.25 DOLBY INTERNATIONAL AB
  • EP4583103B1 patent drawingFigure 1~2
  • EP4583103B1 patent drawingFigure 3~4
  • EP4583103B1 patent drawingFigure 5~6

AI summary

This disclosure falls into the field of audio coding, in particular it is related to the field of providing a framework for providing loudness consistency among differing audio output signals. In particular, the disclosure relates to methods, computer program products and apparatus for encoding and decoding of audio data bitstreams in order to attain a desired loudness level of an output audio signal.