CELP Audio Bandwidth Extension With Low-Delay Excitation Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio signal processing technologies, such as ITU-T G.718 and G.729.1 compliant speech coders, face significant algorithmic delays when coding audio signal bands beyond the core CELP codec bandwidth, as they require sequential decoding and MDCT analysis, leading to inefficiencies in bandwidth extension.
Innovation Solution
The proposed solution involves up-sampling the CELP excitation signal, applying non-linear operations, and using bandpass filtering to extend the audio bandwidth, while synchronizing adaptive codebooks and minimizing alignment errors, thereby reducing algorithmic delays and improving bandwidth extension efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If sequential decoding and MDCT analysis are used to code audio signal bands beyond core CELP codec bandwidth, then bandwidth extension is achieved, but algorithmic delays increase significantly
Solution Approach 1:
The audio signal is divided into two independent processing paths: core CELP codec processing for base bandwidth and separate bandwidth extension processing for higher frequency bands. The bandwidth extension path processes the original input signal directly without requiring sequential decoding, thereby reducing algorithmic delay while achieving bandwidth extension.
Solution Approach 2:
The bandwidth extension processing is performed in parallel with the core CELP encoding, using the original input signal that is already available. This preliminary action eliminates the need to wait for sequential decoding completion, thus reducing the overall algorithmic delay while maintaining bandwidth extension capability.
2Manufacturing precision
If CELP encoded signal is decoded and analyzed to derive difference signal for bandwidth extension, then spectral domain coding is achieved, but processing complexity and delay increase
Solution Approach 1:
The patent extracts only the essential spectral information needed for bandwidth extension from the original input signal, rather than performing full sequential decoding and analysis. This extraction approach maintains spectral domain coding precision while significantly reducing processing complexity by avoiding unnecessary intermediate steps.
Solution Approach 2:
Instead of performing complete sequential decoding and full MDCT analysis, the patent applies partial processing directly to the original signal to derive the necessary difference signal. This partial action achieves sufficient spectral domain coding precision with reduced computational complexity.
Data Source
AI summary
A method for decoding an audio signal having a bandwidth that extends beyond a bandwidth of a CELP excitation signal in an audio decoder including a CELP-based decoder element. The method includes obtaining a second excitation signal having an audio bandwidth extending beyond the audio bandwidth of the CELP excitation signal, obtaining a set of signals by filtering the second excitation signal with a set of bandpass filters, scaling the set of signals using a set of energy-based parameters, and obtaining a composite output signal by combining the scaled set of signals with a signal based on the audio signal decoded by the CELP-based decoder element.