Stereo Extension Apparatus Using MDCT and GMM for Mono Signal Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for converting mono sound signals into stereo signals, such as 'Artificial stereo extension of speech based on inter-channel coherence', often result in stereo signals that differ from real signals due to variations in interchannel coherence, failing to provide user satisfaction.
Innovation Solution
A stereo extension apparatus and method utilizing a database for Gaussian mixture model or hidden Markov model training, combined with modified discrete cosine transform (MDCT) and inverse MDCT processes, to estimate and generate a side signal based on the sum and difference between the mono signal and the estimated side signal, improving sound quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional interchannel coherence method is used to extend mono signal to stereo signal, then the process is simple, but the obtained stereo signal differs from real signal due to ICC variation
Solution Approach 1:
The patent changes the approach from using interchannel coherence (ICC) to using interchannel cross-correlation (ICC) as the key parameter for stereo signal generation. This parameter change allows the system to maintain simplicity while improving accuracy, as ICC better captures the relationship between mid and side signals across different audio conditions and contexts.
Solution Approach 2:
The patent replaces the conventional ICC-based mechanical processing system with a new system based on MDCT transformation and ICC calculation. This substitution introduces a mathematical transformation layer that enables more accurate stereo signal generation while maintaining computational efficiency through standardized signal processing operations.
2Reliability
If a stereo signal is generated from a mono signal, then user satisfaction is improved, but the electronic device requires more complicated processing
Solution Approach 1:
The patent segments the stereo signal generation process into distinct modular stages: MDCT transformation of the mono signal, ICC calculation based on transformed coefficients, and inverse MDCT transformation to generate the final stereo signal. This segmentation allows each module to be independently optimized and implemented, reducing overall device complexity while maintaining high user satisfaction.
Solution Approach 2:
The patent introduces MDCT transformed coefficients as an intermediary representation between the input mono signal and the output stereo signal. This intermediary form enables accurate ICC calculation and facilitates the generation of realistic stereo signals while keeping the processing architecture manageable and implementable in electronic devices.
3Manufacturing precision
If MDCT transformation is applied to mono signal for stereo extension, then sound quality is improved, but processing time and computational load increase
Solution Approach 1:
The patent applies MDCT transformation with periodic windowing functions that exploit the temporal redundancy in audio signals. By using overlapping analysis windows and the inherent periodicity in speech and music signals, the system achieves high-quality stereo extension while reducing effective processing time through efficient reuse of transformed coefficients across adjacent time frames.
Data Source
AI summary
Disclosed herein are a stereo extension apparatus and method. The apparatus includes a database that stores predetermined information as a result of Gaussian mixture model (GMM) training or hidden Markov model (HMM) training; a modified discrete cosine transform (MDCT) transformer that transforms a mono signal through MDCT, a feature parameter extractor that extracts a feature parameter of the mono signal from an MDCT coefficient output from the MDCT transformer, a side signal energy estimator that estimates subband energy of a side signal with reference to information stored in the database based on the feature parameter; an energy controller that obtains the MDCT coefficient of a side signal estimated from the subband energy of the estimated side signal, an inverse MDCT transformer that obtains an estimated side signal by transforming the MDCT coefficient of the estimated side signal through inverse MDCT.