Binaural Audio Decoding Without QMF Multi-Channel Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for decoding multi-channel signals into 2-channel binaural signals require complex operations, including reconstruction in the quadrature mirror filter (QMF) domain and dedicated binaural processing, which increases decoding complexity and hardware requirements, especially in devices with limited resources.
Innovation Solution
A method and device that decode multi-channel signals into 2-channel binaural signals by calculating full band or sub-band channel levels from channel level differences, localizing data based on these levels, and applying weighted head related transfer functions to synthesize 2-channel signals without reconstructing in the QMF domain, thereby simplifying the decoding process and reducing hardware needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multi-channel signals are reconstructed in QMF domain and down-mixed using HRTFs, then binaural output quality is improved, but decoding complexity and hardware requirements increase
Solution Approach 1:
The patent merges the QMF-based multi-channel reconstruction process and the HRTF-based binaural down-mixing process into a single integrated decoding operation. Instead of performing reconstruction then down-mixing as separate stages, the decoder directly synthesizes binaural signals from the compressed mono/stereo input by combining spatial cue processing and HRTF application in one computational pass, thereby maintaining binaural quality while reducing decoding complexity
Solution Approach 2:
The patent creates a universal decoding framework that can handle multiple output formats (binaural, multi-channel, stereo, mono) from a single compressed bitstream through one decoding operation. The decoder is designed to perform spatial rendering for different configurations simultaneously, eliminating the need for separate reconstruction and down-mixing processing chains, thus reducing hardware requirements while maintaining output quality
2Manufacturing precision
If separate operations of reconstructing multi-channel signals and down-mixing using HRTFs are performed, then spatial audio quality is improved, but the number of operations and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing HRTF impulse responses and spatial cue parameters during system initialization or playback preparation. These pre-computed values are then directly applied during decoding without requiring real-time QMF reconstruction, significantly reducing the number of operations needed during actual playback while preserving spatial audio quality through the pre-prepared spatial transformation data
3Measurement precision
If dedicated binaural processing is performed after multi-channel reconstruction, then binaural rendering accuracy is improved, but hardware resources required increase
Solution Approach 1:
The patent merges the binaural rendering function directly into the core decoding process. The spatial rendering unit is integrated with the frequency domain processing blocks, allowing HRTF-based binaural synthesis to be performed as part of the same computational framework used for multi-channel reconstruction. This integration eliminates the need for separate dedicated binaural processing hardware while maintaining rendering accuracy through unified signal processing paths
Data Source
AI summary
A decoding method, medium, and device decoding an input signal, including compressed multi-channel signals as a mono or stereo signal, into 2-channel binaural signals. A full band channel level of each channel in the multi-channel system is calculated from channel level differences between the channels, and data of each channel included in the input signal is localized in directions corresponding to the channels based on the calculated full band channel levels of the channels. Accordingly, the input signal can be output as the 2-channel binaural signals by using simple operations without having to reconstruct multi-channel signals from the input signal in a quadrature mirror filter (QMF) domain.