Gaussian Mixture Model Based Methods for Virtual Microphone Signal Synthesis

摘要

Multichannel audio can immerse a group of listeners in a seamless aural environment. However, severalissues must be addressed, such as the excessive transmission requirements of multichannel audio, as well asthe fact that to-date only a handful of music recordings have been made with multiple channels. Previously,we proposed a system capable of synthesizing the multiple channels of a virtual multichannel recording froma smaller set of reference recordings. In this paper these methods are extended to provide a more generalcoverage of the problem. The emphasis here is on time-varying filtering techniques that can be used toenhance particular instruments in the recording, which is desired in order to simulate virtual microphonesin several locations close and around the sound source.