Audio Masking Model for Accurate Bit Allocation at Dynamic Signal Levels

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio encoding technologies fail to accurately estimate masking thresholds based on known properties of human hearing, leading to inefficient bit allocation and reduced audio quality, particularly for signals with dynamic level changes.

Innovation Solution

A masking model that calculates masking thresholds using energy values and hearing thresholds in quiet, combined with sensitivity values to better emulate human hearing, allowing for more accurate bit allocation and improved audio quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If traditional psychoacoustic models are used to estimate masking thresholds, then the encoding process can be simplified, but the accuracy of bit allocation and audio quality deteriorate

Engineering Contradiction:
Improveencoding process complexityVSAvoidmasking threshold estimation accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent changes the parameters used in masking threshold estimation from traditional psychoacoustic models to a simplified model based on energy values and hearing thresholds in quiet. This parameter change reduces encoding complexity while improving accuracy for signals with dynamic level changes, as the new model directly uses measurable energy values rather than complex psychoacoustic calculations.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces complex psychoacoustic modeling mechanisms with a simpler physical measurement-based approach. Instead of using complex auditory model calculations, the system uses direct measurements of energy values and hearing thresholds in quiet, substituting a simpler measurement-based mechanism for a complex theoretical model.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If masking threshold estimates exceed the JND noise level, then fewer bits are allocated, but audio quality deteriorates due to insufficient bit allocation

Engineering Contradiction:
Improvecompression efficiencyVSAvoidaudio quality
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the masking threshold estimation is continuously refined based on the actual energy values and hearing thresholds measured from the audio signal. This feedback loop ensures that bit allocation is optimized, preventing both over-allocation (which wastes compression efficiency) and under-allocation (which degrades audio quality), by adjusting the masking threshold estimate to match the actual signal characteristics.

Inventive Principle:
Principle #23Feedback

3Reliability

If masking threshold estimates are below the JND noise level, then more bits are allocated, but compression efficiency decreases due to unnecessary bit allocation

Engineering Contradiction:
Improveaudio qualityVSAvoidcompression efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent introduces dynamics into the bit allocation process by making the masking threshold estimate adaptive to the actual signal energy levels and hearing thresholds. Rather than using a fixed or overly conservative threshold, the system dynamically adjusts the masking threshold based on real-time measurements of energy values and hearing thresholds, allowing optimal bit allocation that varies with signal characteristics to maximize both quality and compression efficiency.

Inventive Principle:
Principle #15Dynamics

4Ease of manufacture

If heuristic rules are used in masking threshold models, then model development is simplified, but the accuracy based on known properties of human hearing deteriorates

Engineering Contradiction:
Improvemodel developmentVSAvoidaccuracy based on human hearing properties
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent segments the masking threshold estimation process into distinct, measurable components: energy value calculation and hearing threshold in quiet determination. This segmentation allows each component to be independently optimized and measured against known properties of human hearing, replacing vague heuristic rules with discrete, measurable steps that can be directly validated against auditory perception data.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4726715A2A psychoacoustic model for audio processing
Publication Date: 2026.04.15 DOLBY LABORATORIES LICENSING CORP
  • EP4726715A2 patent drawingFigure 1
  • EP4726715A2 patent drawingFigure 2
  • EP4726715A2 patent drawingFigure 3

AI summary

The present disclosure relates to the field of audio coding, in particular, it relates to a method for encoding audio signals through a masking model based on a hearing threshold of frequency intervals of the audio signal and a measured energy of the audio signal for the corresponding frequency intervals. The disclosure further relates to an encoder that is capable of carrying out the audio encoding method.