Audio Cross-Over Filtering for Low-Bitrate Bandwidth Extension
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio codecs face limitations in bandwidth extension techniques, particularly at low bitrates, leading to loss of high-frequency detail and timbre due to the inability to waveform code perceptually important content above a given cross-over frequency, and introduce artifacts like dissonance and warbling, especially in transform-based implementations like MDCT.
Innovation Solution
The proposed solution involves a cross-over filter that performs frequency-wise weighted addition of decoded core signal and spectral tiles using fade-out and fade-in filters, adapting frequency borders based on signal analysis to minimize ringing artifacts and preserve tonal components, thereby reducing dissonance and warbling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If bandwidth extension techniques are used to enable low bitrate audio coding, then audio compression efficiency is improved, but high-frequency detail and timbre are lost
Solution Approach 1:
The audio spectrum is segmented into low-frequency and high-frequency regions, with different coding strategies applied to each. The low-frequency region is waveform coded to preserve detail, while the high-frequency region uses parametric coding for compression, allowing efficient bitrate allocation that maintains both compression and high-frequency quality.
Solution Approach 2:
Different coding qualities are applied locally to different frequency regions. The low-frequency region receives higher coding quality (waveform coding) to preserve perceptually important content, while the high-frequency region uses lower quality parametric coding, optimizing overall perceptual quality at low bitrates.
2Productivity
If spectral patching is used to fill high-frequency regions, then bandwidth extension is achieved, but artifacts like dissonance and warbling are introduced
Solution Approach 1:
The system dynamically adapts the cross-over frequency between low-frequency and high-frequency regions based on signal analysis. This dynamic adjustment prevents fixed boundary artifacts like dissonance and warbling by optimizing the transition point according to the actual signal characteristics.
Solution Approach 2:
A cross-over filter acts as an intermediary between the low-frequency and high-frequency regions, providing a smooth transition that eliminates abrupt spectral discontinuities. This intermediary filtering prevents the introduction of dissonance and warbling artifacts at the boundary between coded and synthesized regions.
3Productivity
If transform-based bandwidth extension is used, then compression efficiency increases, but ringing artifacts are introduced
Solution Approach 1:
The system converts the potential harm of transform-based processing by using the transform domain for efficient parametric coding while applying inverse filtering and cross-over filtering to eliminate the resulting ringing artifacts. The harmful transform effects are thus converted into beneficial compression efficiency.
Solution Approach 2:
The system changes parameters such as the cross-over frequency and filter characteristics based on signal analysis to minimize ringing artifacts. By adapting these parameters to the signal content, the system maintains compression efficiency while reducing transform-induced artifacts.
4Device complexity
If a fixed cross-over frequency is used for bandwidth extension, then implementation is simplified, but signal adaptability is reduced
Solution Approach 1:
The cross-over frequency is made dynamic rather than fixed, adapting to the signal characteristics through analysis. This dynamic approach improves signal adaptability while maintaining reasonable implementation complexity through efficient signal analysis and adaptive filtering techniques.
Solution Approach 2:
The system performs self-adjustment by analyzing its own signal content and automatically optimizing the cross-over frequency and filter parameters. This self-service capability improves signal adaptability without requiring external control or complex manual configuration.
Data Source
AI summary
Apparatus for decoding an encoded audio signal including an encoded core signal, including: a core decoder for decoding the encoded core signal to obtain a decoded core signal; a tile generator for generating one or more spectral tiles having frequencies not included in the decoded core signal using a spectral portion of the decoded core signal; and a cross-over filter for spectrally cross-over filtering the decoded core signal and a first frequency tile having frequencies extending from a gap filling frequency to an upper border frequency or for spectrally cross-over filtering a first frequency tile and a second frequency tile.