QMF Binaural Decoding for Compressed Multi-Channel Audio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for generating binaural sounds from encoded multi-channel signals are complex and require separate operations for reconstructing multi-channel signals and transforming them into the frequency domain, making them difficult to implement in devices with limited hardware resources, such as mobile audio devices.
Innovation Solution
A decoding method and system that reconstructs multi-channel signals from an encoded input signal in the quadrature mirror filter (QMF) domain, transforms head-related transfer functions (HRTFs) into spatial parameters in the QMF domain, and localizes the signals to generate binaural signals using simple operations without additional processing steps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If separate operations are performed for reconstructing multi-channel signals and transforming them into frequency domain for binaural processing, then binaural sound quality is improved, but device complexity increases and hardware resources are excessive
Solution Approach 1:
The patent combines the signal reconstruction operation and binaural processing operation into a single integrated process. Instead of separately reconstructing multi-channel signals and then transforming them to frequency domain for HRTF processing, the invention performs both operations simultaneously in the QMF domain, reducing the number of processing stages and simplifying the overall system architecture.
Solution Approach 2:
The patent makes the QMF domain processing capable of performing multiple functions: both signal reconstruction and binaural processing with HRTF application. This multi-functional approach eliminates the need for separate frequency domain transformation steps, as the QMF domain itself supports both operations that were previously required in different domains.
2Manufacturing precision
If separate operations are performed for reconstructing multi-channel signals and transforming them into frequency domain, then signal processing accuracy is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary transformation of HRTF data into the QMF domain in advance, so that when signal reconstruction and binaural processing are performed, the HRTF data is already in the appropriate domain for direct application. This preliminary preparation eliminates the need for real-time frequency domain transformation during the main processing flow, reducing processing time.
Solution Approach 2:
By merging the reconstruction and binaural processing operations into a single simultaneous process in the QMF domain, the patent eliminates the sequential execution of separate operations, thereby reducing the total processing time while maintaining signal processing accuracy through the unified approach.
3Adaptability or versatility
If additional processing steps are added for frequency domain transformation and HRTF processing, then binaural processing capability is improved, but ease of manufacture deteriorates
Solution Approach 1:
The patent creates a universal processing framework in the QMF domain that handles both signal reconstruction and binaural processing with HRTF application. This multi-functional capability is built into the core processing architecture, making it suitable for implementation in resource-constrained mobile devices without requiring additional separate processing hardware or complex multi-domain transformations.
Solution Approach 2:
The patent extracts the essential binaural processing functionality and integrates it directly into the QMF domain reconstruction process. By taking out the frequency domain transformation step as a separate operation and incorporating HRTF processing directly into the QMF domain operations, the invention simplifies the implementation while retaining full binaural processing capability.
Data Source
AI summary
A decoding method, medium, and system decoding an input compressed multi-channel signal, as a mono or stereo signal, into 2-channel binaural signals. Channel signals making up the multi-channel signals may be reconstructed from the input compressed signal in the quadrature mirror filter (QMF) domain, and head related transfer functions (HRTFs) for localizing channel signals in the frequency domain, represented as values in the time domain, may be transformed into spatial parameters in the QMF domain. Accordingly, channel signals may be localized in the QMF domain in directions corresponding to the channels, thereby decoding the input compressed signal as 2-channel binaural signals.