Spectral Audio Patching for Adaptive High-Frequency Reconstruction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing bandwidth extension methods in audio coding, such as spectral band replication (SBR), lack flexibility and adaptability, leading to blocking artifacts and inefficient processing due to the use of a single patching algorithm that does not account for signal characteristics or domain transitions.

Innovation Solution

Implementing a system that switches between multiple spectral domain patching algorithms, including harmonic transposition, non-harmonic copying-up SBR, and non-linear distortion, within the spectral domain to generate a modified spectral representation, allowing for improved perceptual quality and reduced complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a single patching algorithm is used in existing bandwidth extension methods, then the processing is simpler, but the adaptability to different signal characteristics is poor and blocking artifacts occur

Engineering Contradiction:

Improveadaptability to signal characteristicsVSAvoidprocessing complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system dynamically selects from multiple patching algorithms (harmonic transposition, non-harmonic copying-up SBR, non-linear distortion) based on signal characteristics such as tonality and transient content. This dynamic adaptation allows the system to optimize processing for different signal types while maintaining manageable complexity through automated selection criteria.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes processing parameters by selecting different patching algorithms based on analyzed signal characteristics. The encoder analyzes tonality, transient content, and spectral properties to determine which algorithm parameters (harmonic ratios, copying factors, distortion levels) are most appropriate for the current signal segment.

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If domain transitions are made in existing methods, then different processing approaches can be applied, but blocking artifacts are introduced

Engineering Contradiction:

Improveprocessing flexibilityVSAvoidblocking artifacts

Core Design Contradiction:

Adaptability or versatilityVSObject-generated harmful factors

Solution Approach 1:

The system maintains continuous processing within the spectral domain throughout the bandwidth extension process, avoiding transitions to other domains. This continuous spectral domain processing eliminates the blocking artifacts that would otherwise be introduced by domain transitions, while still providing processing flexibility through algorithm selection.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The spectral domain serves as a consistent intermediary representation throughout the entire processing chain. By keeping all operations (patching, bandwidth extension, synthesis) in the spectral domain, the system avoids the harmful effects of domain transitions while maintaining the ability to apply different processing approaches through algorithm selection.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Manufacturing precision

If bandwidth extension is applied to maintain audio quality at low bit rates, then perceptual quality is improved, but processing complexity increases

Engineering Contradiction:

Improveperceptual qualityVSAvoidprocessing complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The bandwidth extension process is segmented into distinct operational phases: spectral analysis, algorithm selection based on signal characteristics, patching operation selection (harmonic, non-harmonic, or distortion-based), and synthesis. This segmentation allows the system to apply complexity only where needed while maintaining efficiency in other stages.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs self-service by automatically analyzing signal characteristics and selecting the most appropriate patching algorithm without external intervention. The encoder autonomously determines tonality, transient content, and spectral properties to guide algorithm selection, reducing the need for complex external control mechanisms while maintaining high perceptual quality.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUSRE50692E1Apparatus and method for generating a synthesis audio signal and for encoding an audio signal

Publication Date: 2025.12.09 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV

AI summary

An apparatus for generating a synthesis audio signal using a patching control signal has a first converter, a spectral domain patch generator, a high frequency reconstruction manipulator and a combiner. The first converter is configured for convert- ing a time portion of an audio signal into a spectral represen- tation. The spectral domain patch generator is configured for performing a plurality of different spectral domain patching algorithms, wherein each patching algorithm generates a modified spectral representation having spectral components in an upper frequency band derived from corresponding spec- tral components in a core frequency band of the audio signal. The spectral domain patch generator is furthermore config- ured to select a first spectral domain patching algorithm from the plurality of patching algorithms for a first time portion and a second spectral domain patching algorithm from the plural- ity of patching algorithm for a second different time portion in accordance with the patching control signal to obtain the modified spectral representation.