Loudness Metadata Conversion for Consistent Audio Playback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing multimedia devices face challenges in providing consistent audio leveling across varying listening environments and diverse audio content formats, with limitations in computational complexity and codec-agnostic applicability, particularly in portable media players.

Innovation Solution

A method utilizing encoder-generated metadata for dynamic range control, converting input audio signals into output signals with a target output level by applying gain values derived from reversible relationships between different loudness measurement procedures, such as ITU-R BS.1770 and Replay Gain, to ensure consistent audio playback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple loudness measurement procedures are implemented to support different audio formats, then compatibility across codecs is improved, but computational complexity increases

Engineering Contradiction:
Improvecodec compatibilityVSAvoidcomputational complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent combines multiple loudness measurement procedures (ITU-R BS.1770 and Replay Gain) into a unified metadata structure. The encoder generates both types of loudness values and stores them together with the audio content, allowing the decoder to use the appropriate measurement type based on the audio format without implementing separate complex processing paths for each codec.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces a standardized metadata structure as an intermediary between the encoder and decoder. This metadata carries loudness information from both ITU-R BS.1770 and Replay Gain procedures, serving as a bridge that allows different audio formats to be handled uniformly without requiring the decoder to implement multiple complex measurement algorithms.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If loudness normalization is applied to all audio content, then consistent playback level is achieved, but dynamic range is reduced

Engineering Contradiction:
Improveaudio leveling consistencyVSAvoiddynamic range reduction
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent implements dynamic range control with adjustable parameters that allow the system to adapt between different operating modes. The encoder can set compression parameters, and the decoder can adjust the balance between normalization strength and dynamic range preservation based on the listening environment and content type, making the system flexible rather than fixed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent uses parameter-based control where the encoder specifies compression parameters in the metadata, and the decoder adjusts normalization behavior based on these parameters. By changing parameters like compression ratio and threshold, the system can achieve consistent playback levels while preserving dynamic range for content where this is important.

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If encoder-generated metadata is used for loudness control, then computational complexity at decode is reduced, but encoding complexity increases

Engineering Contradiction:
Improvedecoder complexityVSAvoidencoding complexity
Core Design Contradiction:
Device complexityVSEase of manufacture

Solution Approach 1:

The patent performs loudness measurement and metadata generation during the encoding phase, before the audio content is stored or transmitted. The encoder calculates both ITU-R BS.1770 and Replay Gain loudness values and embeds them in the metadata, so that the decoder only needs to read and apply the pre-computed values without performing complex loudness measurements itself.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3236586B1System for combining loudness measurements in a single playback mode
Publication Date: 2019.11.27 DOLBY INTERNATIONAL AB
  • EP3236586B1 patent drawingFigure 1a~1b
  • EP3236586B1 patent drawingFigure 2a~2b
  • EP3236586B1 patent drawingFigure 2c~2d

AI summary

The present document relates to processing of multimedia data, notably the encoding, the transmission, the decoding and the rendering of multimedia data, e.g. audio files or bitstreams. In particular, the present document relates to the implementation of loudness control in multimedia players. A method for providing loudness related data to a media player is described. The method comprises the steps of providing a first loudness related value associated with an audio signal; wherein the first loudness related value has been determined according to a first procedure; of converting the first loudness related value into a second loudness related value using a reversible relation; wherein the second loudness related value is associated with a second procedure for determining loudness related values; of storing the second loudness related value in metadata associated with the audio signal; and of providing the metadata to the media player.