Extracting and Modifying the Spatial Information in Stereo Audio

Abstract

In this paper, a method to extract the spatial information and signals in stereo, and then synthesis the new sound field is proposed. The objective is to synthesize appropriate sound field corresponding to different listening conditions. The discussed situation is limited to the recording of chorus, multi-sources playing the same melody by the same music instruments and being aligned in line. By assuming that the frequency magnitudes of spectrogram of each source are similar and the procedure of signals received by microphone is similar to image corrupt of liner distortion, we apply the concept of image restoration to extract the spatial information and single source. The simulation is performed to confirm the method and the results demonstrate that a sound field similar to the original sound field can be synthesized using the extracted single source and spatial information. Also the spatial information can be modified to synthesize a different sound field.