Audio Coding Scheme Selection Using Linear Prediction Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio coding methods face high operational complexity and overheads due to the need for closed-loop mode selection of sub-coders, which is inefficient in dynamically selecting the most suitable coding scheme for mixed audio signals containing both speech and non-speech elements.

Innovation Solution

An audio coding method that estimates the reference linear prediction efficiency of an audio frame to determine a matching audio coding scheme, reducing the need for complete coding operations and thereby lowering operational complexity and overheads by selecting between linear prediction-based and non-linear prediction-based coding schemes based on efficiency thresholds.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If closed-loop mode selection is used to dynamically select the most suitable sub-coder, then coding quality is improved, but operational complexity and overheads increase

Engineering Contradiction:
Improvecoding qualityVSAvoidoperational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts audio features (temporal and spectral) in advance before actual coding decision-making. This preliminary feature extraction enables the system to predict suitable coding modes without performing complete closed-loop coding operations, thereby reducing operational complexity while maintaining coding quality through informed mode selection.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediate feature extraction and analysis stage that acts as a mediator between the raw audio signal and the final coding mode selection. By using extracted features as intermediaries to evaluate coding suitability, the system avoids direct closed-loop comparisons of multiple sub-coders, reducing computational overhead while preserving coding quality decisions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If closed-loop mode selection is used to compare quality of coded frames, then sub-coder selection accuracy is improved, but overheads of audio coding increase

Engineering Contradiction:
Improvesub-coder selection accuracyVSAvoidoverheads of audio coding
Core Design Contradiction:
Measurement precisionVSLoss of substance

Solution Approach 1:

The patent extracts and utilizes specific features (temporal and spectral) from the audio signal to determine coding mode suitability. By taking out and analyzing these key features separately, the system achieves accurate sub-coder selection without performing complete coding operations for each candidate, thereby reducing the overheads associated with closed-loop comparisons.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of performing complete coding operations for each sub-coder candidate (excessive action), the patent performs partial analysis by extracting and evaluating key audio features. This partial action is sufficient to make accurate sub-coder selection decisions while significantly reducing the computational overheads of full closed-loop comparisons.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If each sub-coder codes the input current audio frame once for comparison, then coding quality assessment is improved, but coding operation complexity increases

Engineering Contradiction:
Improvecoding quality assessmentVSAvoidcoding operation efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent performs preliminary feature extraction (temporal and spectral characteristics) before the actual coding quality assessment. This preliminary action provides the necessary information for evaluating sub-coder suitability without requiring each sub-coder to completely code the audio frame, thus maintaining assessment accuracy while improving coding operation efficiency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses extracted audio features as intermediaries to assess coding quality suitability. Instead of requiring complete coding operations by each sub-coder, the system uses these intermediate feature representations to evaluate and select the most appropriate coding mode, thereby improving operational efficiency while maintaining assessment precision.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3618069B1Audio coding method and related apparatus
Publication Date: 2024.03.20 HUAWEI TECH CO LTD
  • EP3618069B1 patent drawingFigure 1
  • EP3618069B1 patent drawingFigure 2
  • EP3618069B1 patent drawingFigure 3-a~3-b

AI summary

An audio coding method and a related apparatus are disclosed. The audio coding method includes: estimating reference linear prediction efficiency of a current audio frame; determining an audio coding scheme that matches the reference linear prediction efficiency of the foregoing current audio frame; and performing audio coding on the foregoing current audio frame according to the audio coding scheme that matches the reference linear prediction efficiency of the foregoing current audio frame. The technical solutions provided in embodiments of the present invention help reduce overheads of audio coding.