Multi-channel Audio Playback Spatial Reproduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in converting multi-microphone captured signals into high-quality binaural signals that retain the original spatial representation, especially when using omnidirectional microphones placed in various configurations, failing to provide an authentic listening experience similar to recordings made with an artificial head.
Innovation Solution
The system employs directional analysis and binaural synthesis techniques, using three omnidirectional microphones to estimate sound source directions and introduce an ambience component, creating coherent downmix signals that can be rendered as binaural or multi-channel audio, allowing for flexible playback on different speaker setups and head tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If multiple omnidirectional microphones are used to capture audio signals, then recording flexibility and ease of operation are improved, but the spatial representation quality and manufacturing precision deteriorate
Solution Approach 1:
The patent introduces an artificial head model as an intermediary between the multiple omnidirectional microphones and the final binaural output. The transfer functions representing the artificial head's acoustic paths serve as mediators that transform the multi-microphone signals into authentic binaural signals, thereby resolving the spatial representation quality issue while maintaining recording flexibility
Solution Approach 2:
The patent creates a virtual copy of the artificial head recording system by using multiple omnidirectional microphones with known positions. Through computational processing involving transfer functions and directional analysis, it reproduces the binaural effect that would otherwise require physical artificial head microphones, thus maintaining spatial accuracy without sacrificing operational flexibility
2Manufacturing precision
If binaural synthesis techniques are applied to multi-microphone signals, then spatial representation quality is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary computation of transfer functions between each microphone position and the artificial head ear positions before actual signal processing. These pre-computed transfer functions are stored and reused during playback, significantly reducing the real-time computational complexity while maintaining high spatial representation quality
Solution Approach 2:
The patent divides the binaural synthesis process into distinct segments: directional analysis of incoming signals, separation of direct sound from reverberation, application of appropriate transfer functions for different sound sources, and combination of processed signals. This segmentation makes the complex processing more manageable and implementable
3Measurement precision
If directional analysis and subband processing are used to emphasize dominant sound sources, then spatial accuracy is improved, but loss of information in ambient components increases
Solution Approach 1:
The patent applies different processing qualities to different frequency subbands and spatial components. Dominant sound sources in specific subbands receive enhanced directional processing for accurate localization, while ambient components in other subbands are preserved with minimal processing. This local differentiation maintains spatial accuracy for important sources while preserving ambient information
Data Source
AI summary
Techniques are presented for creating multichannel output signals from input audio signals. A first signal is determined based on a number of subbands into which the input audio signals are divided and based at least in part on a directional estimation wherein the subbands having dominant sound source directions are emphasized relative to subbands having directional estimates that deviate from directional estimates of the dominant sound source directions. A second signal is determined based on the number of subbands wherein an ambient component is introduced to create a perception of an externalization for a sound image. A resultant audio signal is created using the first and second signals. The resultant audio signal is one of a number of multichannel signals. Additionally, it is determined whether binaural audio output or multichannel audio output (or both) is to be output, and the appropriate number of audio output signals are determined and output.