Audio Encoding Device Using Angle Estimation for Ambisonic B-Format
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating Ambisonic B-format sound signals in virtual reality applications require a large number of microphones and suffer from spatial aliasing and signal-to-noise ratio issues, making them unsuitable for mobile devices.
Innovation Solution
An audio encoding device and method that uses a low number of microphones (N≥3) to estimate angles of incidence and derive A-format direct sound signals, which are then transformed into Ambisonic B-format signals using a transformation matrix, while also incorporating short-time Fourier transformations and de-correlation filters to enhance sound quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If differential microphone arrays with delay and adding beam-forming are used to generate Ambisonic B-format signals, then first order virtual microphone signals can be derived, but spatial aliasing occurs which reduces the bandwidth to frequencies f in the range f < c/(2*dmic)
Solution Approach 1:
The patent changes the processing parameters by applying frequency-dependent weighting factors to different microphone pairs based on their estimated angles of incidence. This allows the system to adaptively select and weight microphone pairs according to the frequency content and spatial distribution of sound sources, thereby extending the effective bandwidth beyond the traditional spatial aliasing limit while maintaining measurement precision.
2Measurement precision
If a dense distribution of microphones is used to sample the sound field for converting to spherical harmonics, then B-format signals can be generated, but the required number of microphones is too high for consumer applications
Solution Approach 1:
The patent creates virtual microphone signals through mathematical processing rather than using physical microphones for every sampling point. By estimating angles of incidence from a small number of physical microphones and synthesizing A-format signals that represent what additional microphones would capture, the system achieves sound field sampling accuracy equivalent to a dense microphone distribution while using only a few physical microphones.
Solution Approach 2:
The patent replaces the mechanical approach of using many physical microphones to sample the sound field with a computational approach. Instead of physically placing microphones at multiple positions to capture spherical harmonics, the system uses signal processing techniques (angle estimation, beamforming, spherical harmonic transformation) to synthesize the equivalent information from a minimal microphone array.
3Device complexity
If only a few microphones are used for consumer applications, then hardware requirements are reduced, but linear processing leads to signal to noise ratio issues at low frequencies and aliasing at high frequencies
Solution Approach 1:
The patent implements dynamic processing by continuously estimating angles of incidence and adaptively weighting microphone pair contributions based on current acoustic conditions. The system dynamically adjusts which microphone pairs are used and with what weightings, allowing it to maintain high signal-to-noise ratio across a wide frequency range despite using only a few microphones. This dynamic adaptation prevents both low-frequency noise issues and high-frequency aliasing that plague static linear processing approaches.
Data Source
AI summary
A method and a device encode N audio signals, from N microphones where N≥3. For each pair of the N audio signals an angle of incidence of direct sound is estimated. A-format direct sound signals are derived from the estimated angles of incidence by deriving from each estimated angle an A-format direct sound signal. Each A-format direct sound signal is a first-order virtual microphone signal, for example, a cardioids signal.