In this research, we propose a multichannel-to-loudspeaker array upmixing algorithm. To take advantage of a loudspeaker array, we introduce an approach based on audio source separation. During the analysis phase, multichannel signals are separated into element signals in stereo format through stereo-channel extraction. They are then separated into source signals by the Laplacian Mixture Models of the features from the stereo signals. During the synthesis phase, the sources are rendered as virtual sources through Wave Field Synthesis, or as focused sources. Subjective tests show that the proposed algorithm outperforms the comparison algorithm in terms of localization quality.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.