Stereo Downmixing with Position-Based HRTF Mixing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for downmixing multi-channel surround audio signals to 2-channel stereo signals often result in displaced or obscure sound images due to the application of head-related transfer functions, leading to a poor sound image feeling.
Innovation Solution
An audio signal processing device that synthesizes surround audio signals with adjusted head-related transfer function contributions based on the relative positions of loudspeakers to maintain sound image localization and clarity, using a processor to generate 2-channel stereo signals by applying varying ratios to audio signals convolved with head-related transfer functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If head-related transfer functions are convolved into all surround audio channels at equal ratios, then spatial audio realism is improved, but sound image localization becomes displaced or obscure
Solution Approach 1:
The patent applies different convolution ratios of head-related transfer functions to different audio channels based on their specific spatial positions. Front channels (center, left, right) use a first convolution ratio, while rear channels (left surround, right surround, left back, right back) use a second convolution ratio. This local differentiation maintains spatial realism while preventing sound image displacement.
Solution Approach 2:
The patent changes the convolution ratio parameter of head-related transfer functions based on the spatial position of loudspeakers. By adjusting this parameter differently for front and rear channels, the system optimizes both spatial audio realism and sound image localization accuracy for each specific channel.
2Adaptability or versatility
If head-related transfer functions are applied to downmix surround audio to stereo, then spatial audio immersion is improved, but sound image clarity deteriorates
Solution Approach 1:
The patent applies head-related transfer function convolution selectively based on channel position. Front channels receive convolution at one ratio to maintain clarity, while rear channels receive convolution at a different ratio to provide spatial immersion. This local quality differentiation resolves the contradiction between immersion and clarity.
3Device complexity
If uniform convolution ratios are used for all channels, then processing simplicity is maintained, but sound image localization accuracy deteriorates
Solution Approach 1:
The patent implements local quality by assigning different convolution ratios to different channel groups (front vs. rear). This approach maintains reasonable processing simplicity while significantly improving sound image localization accuracy compared to uniform convolution, as the differentiation is based on clear spatial positioning.
Solution Approach 2:
The patent segments the surround audio channels into different groups (front channels and rear channels) and applies different convolution ratios to each segment. This segmentation enables differentiated processing that improves localization accuracy while keeping the system manageable through clear categorization.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The device effectively maintains sound image localization and prevents displacement, ensuring a good sound image feeling by adjusting head-related transfer function contributions based on loudspeaker positions, thus enhancing the sound image clarity and localization.
Implementation Method 1
convolves head-related transfer functions, which are sound transfer functions from a loudspeaker of each channel to the left and right ears of the listener respectively, into audio signals of that channel
Data Source
AI summary
An audio signal processing device for downmixing surround audio signals so as to obtain a good sound image includes a pseudo surround audio signal generator for generating surround audio signals SigA of 7.1.4 channels from input 2-channel stereo audio signals SigIN, a signal processor for signal-processing the surround audio signals SigA and outputting results as downmixing signals SigB, and a downmixer for synthesizing the downmixing signals SigB and outputting 2-channel stereo audio signals SigOUT. For an input audio signal SigA of each channel, the signal processor synthesizes the audio signal and an audio signal obtained by convolving a head-related transfer function from a loudspeaker corresponding to the audio signal to a listener into the audio signal, at a ratio matching the position of the loudspeaker, and outputs the result as an audio signal SigB.