Audio Decoding Gain Waveform Interpolation for Frame Continuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio coding techniques, such as MPEG AAC, face issues with discontinuous temporal waveforms in audio signals due to uniform gain correction across frames, leading to degradation in sound quality when significant gain changes occur between frames.
Innovation Solution
A decoding apparatus and method that allows for arbitrary gain values to be designated for each sample within a frame, using linear or non-linear interpolation to generate a smooth gain waveform, reducing code amount and improving sound quality by accurately reproducing the original waveform with less data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If uniform gain correction is applied across all samples within one frame, then the code amount is reduced and processing is simplified, but discontinuous temporal waveforms occur between frames causing degradation in sound quality
Solution Approach 1:
The frame is divided into multiple blocks, and gain values are calculated independently for each block rather than uniformly for the entire frame. This segmentation allows different gain values to be applied to different temporal segments, maintaining waveform continuity while managing code amount through selective precision.
Solution Approach 2:
Different gain values are assigned to different blocks within the frame based on local characteristics. The gain calculation adapts to local temporal variations, applying higher precision where needed and lower precision where uniform correction suffices, thus balancing sound quality and code efficiency.
2Reliability
If gain values are designated for each sample within a frame, then temporal waveform continuity is improved and sound quality increases, but the code amount and processing complexity increase
Solution Approach 1:
Instead of processing each sample individually, the frame is segmented into blocks where a single gain value is calculated per block. This reduces the number of gain calculations from thousands of samples to a manageable number of blocks, significantly lowering processing complexity while maintaining waveform continuity.
Solution Approach 2:
Gain values are calculated for representative samples or specific positions within each block rather than all samples. This partial action approach provides sufficient waveform continuity without the excessive computational burden of per-sample processing.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The present technology relates to a decoding apparatus, a decoding method and a program which make it possible to obtain sound with higher quality. A demultiplexing circuit demultiplexes an input code string into a gain code string and a signal code string. A signal decoding circuit decodes the signal code string to output a time series signal. A gain decoding circuit decodes the gain code string. That is, the gain decoding circuit reads out gain values and gain inclination values at predetermined gain sample positions of the time series signal and interpolation mode information. An interpolation processing unit obtains a gain value at each sample position between two gain sample positions through linear interpolation or non-linear interpolation according to the interpolation mode based on the gain values and the gain inclination values. A gain applying circuit adjusts a gain of the time series signal based on the gain values. The present technology can be applied to a decoding apparatus.