Stereo Audio Encoding Cross-Correlation Spatial Image
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional stereo speech coding methods lack sufficient spatial information, particularly inter-channel correlation (ICC), which results in inferior spatial image reproduction in decoded stereo speech signals.
Innovation Solution
A stereo speech coding apparatus that calculates and transmits cross-correlation coefficients between channel signals, using these coefficients to enhance spatial information and improve the decoding process by comparing the original and reconstructed cross-correlation values, thereby incorporating ICC into the decoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional stereo speech coding methods are used, then the coding process is simple, but the spatial image reproduction is inferior due to lack of inter-channel correlation information
Solution Approach 1:
The patent calculates cross-correlation coefficients between channel signals in advance during the coding process, and transmits these coefficients as side information to the decoder. This preliminary calculation and transmission of spatial information enables the decoder to reconstruct accurate spatial images without requiring complex processing during decoding.
Solution Approach 2:
The patent introduces cross-correlation coefficients as an intermediary parameter that bridges the encoder and decoder. These coefficients serve as side information that carries spatial relationship data between channels, allowing the decoder to reconstruct the spatial image without directly transmitting the full spatial information.
2Loss of information
If cross-correlation coefficients are calculated and transmitted, then the spatial image reproduction is improved, but the bit rate increases due to additional side information
Solution Approach 1:
The patent extracts only the essential spatial information in the form of cross-correlation coefficients from the stereo signal, rather than transmitting all spatial data. This selective extraction of critical parameters reduces the amount of side information that needs to be transmitted while still achieving improved spatial image reproduction.
3Measurement precision
If cross-correlation coefficients are used in decoding, then the spatial image quality is enhanced, but the decoding process becomes more complex
Solution Approach 1:
The patent enables the decoder to use the received cross-correlation coefficients to automatically reconstruct the spatial image without requiring complex external processing or additional side information. The decoder leverages these coefficients to generate the spatial relationships between channels through standard decoding operations.
Data Source
AI summary
Disclosed is a stereo audio encoding device capable of improving a spatial image of a decoded audio in stereo audio encoding. In this device, an original cross correlation calculation unit (101) calculates a mutual relationship coefficient (C1) between the original L channel signal and the original R channel signal. A stereo audio reconfiguration unit (104) subjects the inputted L channel signal and the R channel signal to encoding and decoding so as to generate an L channel reconfigured signal (L′) and an R channel reconfigured signal (R′). A reconfiguration cross correlation calculation unit (105) calculates a cross correlation coefficient (C2) between the L channel reconfigured signal (L′) and the R channel reconfigured signal (R′). A cross correlation comparison unit (106) calculates and outputs a comparison result &agr; between the cross correlation coefficient (C1) and the cross correlation coefficient (C2).