Audio Signal Resampling Using Predicted Future Samples for Low Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio frequency signal resampling methods, such as those used in the AMR-WB codec, introduce delays that can lead to audible discontinuities and degrade the quality of the encoded signal, particularly when using FIR filters for resampling, as they require adding zeros to simulate future signal values, which can result in errors and periodic discontinuities.
Innovation Solution
The method involves adaptive linear prediction to determine the number of future signal samples based on a chosen resampling delay, constructing a resampling support vector from current and predicted future samples, and applying a resampling filter to reduce resampling delay while maintaining performance similar to continuous resampling, allowing for switching between coding modes and configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If FIR filter is used for resampling, then filtering characteristics are improved, but delay increases
Solution Approach 1:
The patent applies preliminary action by predicting future signal samples before they are actually available. The linear prediction module generates predicted future samples that are used in the resampling process, allowing the system to act as if the future signal values are already known, thereby reducing the delay caused by traditional FIR filtering while maintaining filtering characteristics.
2Ease of operation
If zeros are added to simulate future signal values, then resampling can be performed, but errors and periodic discontinuities are introduced
Solution Approach 1:
The patent uses copying by creating predicted copies of future signal samples based on the linear prediction model. Instead of using zeros or actual future samples, the system generates predicted sample values that copy the expected signal behavior, thereby maintaining signal quality and avoiding the errors and discontinuities that would result from using zeros.
3Reliability
If resampling delay is reduced, then audio quality is improved, but filtering performance may deteriorate
Solution Approach 1:
The patent substitutes the mechanical FIR filtering system with a linear prediction-based resampling system. Instead of relying on the mechanical convolution process of FIR filters that inherently introduces delay, the system uses linear prediction to directly generate resampled values, thereby reducing delay while maintaining adequate filtering performance through the predictive model.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for resampling an audio frequency signal in an audio frequency signal coding or decoding, characterized in that the method comprises the following steps per block of signal to be resampled: - determination by adaptive linear prediction of a number of future signal samples, this number being defined as a function of a chosen resampling delay; - construction of a resampling support vector from at least the samples of the current signal block and the determined future signal samples; - application of a resampling filter on the samples of the resampling support vector, the method further comprising the following steps: - switching between two coding modes, - resampling the current signal block as a result of the switching.The invention also relates to a resampling device implementing the described process, an encoder and a decoder comprising at least one device.