Sound Prioritisation System for Audio Clips

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current audio reproduction systems in complex entertainment content, such as video games, struggle to prioritize sounds effectively, often omitting important sounds due to subjective importance assignment and context neglect, leading to incomplete sound reproduction and listener interference.

Innovation Solution

A sound prioritization method that analyzes perceptually relevant sound features, considering factors like volume, masking, context, and location, to assign priority values, ensuring that only the most important sounds are reproduced, using machine learning algorithms and feature extraction to generate high-quality audio output.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If manual sound prioritisation is performed based on subjective importance assignment, then the audio reproduction can be simplified, but important sounds may be omitted in preference of less-important sounds

Engineering Contradiction:
Improveaudio reproduction complexityVSAvoidsound prioritisation accuracy
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent replaces manual subjective prioritisation with an automated machine learning system that analyzes audio signals and assigns priority values based on objective acoustic features, thereby eliminating human subjectivity while maintaining system simplicity through automated processing

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

Solution Approach 2:

The system enables audio signals to self-evaluate their own importance through automated analysis of acoustic features such as volume, frequency, and temporal characteristics, without requiring external manual intervention to determine prioritisation

Inventive Principle:
Principle #25Self-service

2Ease of manufacture

If manual sound prioritisation is performed without considering context, then the process can be simplified, but the context in which the sound appears is not reflected in the assigned precedence

Engineering Contradiction:
Improveprioritisation process simplicityVSAvoidcontextual relevance accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system performs preliminary analysis of contextual factors including scene type, character actions, and narrative importance before assigning priority values, ensuring that contextual relevance is captured in advance of the actual sound reproduction decision

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent dynamically adjusts priority assignment parameters based on contextual variables such as scene genre, character hierarchy, and narrative significance, allowing the same sound to receive different priority values depending on its contextual embedding

Inventive Principle:
Principle #35Parameter changes

3Productivity

If a fixed number of top N sounds are reproduced, then the audio output can be controlled, but sounds that interfere with each other are not identified and may obscure important sounds

Engineering Contradiction:
Improveaudio reproduction efficiencyVSAvoidimportant sound omission
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent introduces an intermediary analysis layer that evaluates pairwise interactions between sound signals to identify masking relationships, using acoustic feature comparison to determine which sounds would obscure others before final reproduction decisions are made

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback mechanisms where the prioritisation decisions are continuously refined based on analysis of sound interactions and masking effects, adjusting priority values to ensure that important sounds remain audible even when multiple sounds are reproduced simultaneously

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3780660B1Sound prioritisation system and method
Publication Date: 2023.08.23 SONY INTERACTIVE ENTERTAINMENT LLC
  • EP3780660B1 patent drawingFigure 1
  • EP3780660B1 patent drawingFigure 2~3
  • EP3780660B1 patent drawingFigure 4~6

AI summary

A system for determining prioritisation values for two or more sounds within an audio clip, the system comprising a feature extraction unit operable to extract characteristic features from the two or more sounds, a feature combination unit operable to generate a combined mix comprising extracted features from the two or more sounds, an audio assessment unit operable to identify the contribution of one or more of the features to the combined mix, a feature classification unit operable to assign a saliency score to each of the features in the combined mix, and an audio prioritisation unit operable to determine relative priority values for the two or more sounds in dependence upon the assigned saliency scores for each of one or more features of the sounds.