Selective Harmonic Prediction in Transform-Domain Audio Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio coding methods, such as frequency-domain prediction (FDP), suffer from high computational complexity and limited prediction gain due to the need for backward adaptivity and prediction of noisy components, leading to reconstruction errors.
Innovation Solution
Implement a predictive encoding and decoding scheme that selectively applies prediction to spectral coefficients based on a spacing value, such as the fundamental frequency or its multiples, reducing complexity by applying it only to harmonic signal components, and transmitting prediction errors rather than coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If predictive encoding is applied to all spectral coefficients, then coding efficiency is improved, but computational complexity increases significantly
Solution Approach 1:
The patent divides the spectrum into multiple bands and applies predictive encoding selectively to different bands based on their characteristics. This segmentation allows the system to apply complex predictive encoding only where necessary (in bands with high correlation) while using simpler methods in other bands, thereby resolving the contradiction between coding efficiency and computational complexity.
Solution Approach 2:
The patent implements different encoding strategies for different spectral regions. By analyzing the correlation properties of each band and applying predictive encoding locally only where it provides significant benefit, the system achieves high coding efficiency in critical regions while maintaining overall computational feasibility.
2Manufacturing precision
If complex predictive encoding is applied to improve audio quality, then perceptual quality is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary analysis of spectral correlation properties before applying predictive encoding. By pre-identifying bands that benefit from predictive encoding and preparing prediction models in advance, the system reduces actual encoding processing time while maintaining high audio quality where it matters most.
Solution Approach 2:
The patent applies predictive encoding selectively to only those spectral bands where it provides perceptible quality improvement, rather than applying it uniformly across the entire spectrum. This partial application approach achieves high audio quality in critical regions while significantly reducing overall processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An encoder for encoding an audio signal. The encoder is configured to encode the audio signal in a transform domain or filter-bank domain, wherein the encoder is configured to determine spectral coefficients of the audio signal for a current frame and at least one previous frame, wherein the encoder is configured to selectively apply predictive encoding to a plurality of individual spectral coefficients or groups of spectral coefficients which are separated by at least one spectral coefficient.