Spatial audio array processing system and method
摘要
說明書
Further objects of the present disclosure provide for a spatial audio processing system to overcome deficiencies associated with prior art single channel noise reduction techniques that are ineffective against noises that have similar time-frequency distributions as the target source.
Further objects of the present disclosure provide for a spatial audio processing system to overcome deficiencies associated with prior art multi-channel noise reduction techniques that require noise references constrained to a limited number of situations.
Further objects of the present disclosure provide for a spatial audio processing system to overcome deficiencies associated with prior art multi-channel techniques, such as beamforming and blind source separation, that require a large number of microphones and additional prior knowledge—such as spatial locations of each element in the array of transducers, noise statistics of the current acoustic space, transducer array calibration, target source location(s), and noise source location(s).
Further objects of the present disclosure provide for a spatial audio processing system to overcome deficiencies associated with prior art computational auditory scene analysis techniques that require relatively higher SNR levels or other knowledge, such as when the target talker is speaking.
