Audio Signal Encoding Using Predicted Frame Residuals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audio coding technologies face challenges in achieving better compression ratios for efficiently transmitting audio signals, particularly with the increasing complexity of audio signals.
Innovation Solution
An audio signal encoding method that generates a predicted current frame signal by calculating and synthesizing phase and gain differences between the current and previous frame signals, and outputs a reconstructed residual signal, which is then transmitted to a decoder for decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If audio signals are compressed for efficient transmission, then transmission efficiency is improved, but compression ratio is insufficient
Solution Approach 1:
The patent applies preliminary action by generating a predicted current frame signal in advance using the reconstructed previous frame signal before actual transmission. This prediction is performed beforehand to reduce the amount of data that needs to be transmitted, thereby improving both transmission efficiency and compression ratio simultaneously.
Solution Approach 2:
The patent extracts and transmits only the residual signal (difference between current and predicted frames) rather than transmitting the entire audio signal. This extraction approach significantly reduces the quantity of transmitted data while maintaining audio quality, resolving the contradiction between transmission efficiency and compression ratio.
2Quantity of substance
If difference information between frames is coded, then amount of transmitted information is reduced, but compression efficiency is insufficient
Solution Approach 1:
The patent performs preliminary prediction by generating the predicted current frame signal using temporal correlation with the previous frame before encoding the residual. This preliminary action reduces the amount of information that needs to be transmitted while improving compression efficiency through better exploitation of temporal redundancy.
Solution Approach 2:
The patent changes parameters by transforming the audio signal from time domain to frequency domain using MDCT, and by applying quantization to the residual coefficients. These parameter changes enable more efficient representation of the difference information, reducing transmitted data量 while improving compression efficiency.
Data Source
AI summary
Disclosed are an audio signal encoding/decoding method and an apparatus for performing the same. An audio signal encoding method includes receiving a current frame signal and a reconstructed previous frame signal, generating a predicted current frame signal, based on the current frame signal and the reconstructed previous frame signal, and outputting a reconstructed residual signal, based on the current frame signal and the predicted current frame signal.