We present a real-time sound source separation algorithm which performs the task of source separation based on the lateral displacement of a source within the stereo field. The algorithm exploits the use of the pan pot as a means to achieve image localisation within stereophonic recordings. As such, only an interaural intensity difference exists between left and right channels for a single source. Gain scaling and phase cancellation techniques are used in the frequency domain to expose frequency dependent nulls across the azimuth plane. The position of these nulls in conjunction with magnitude estimation and grouping techniques are then used to resynthesise separated sources. Results obtained from real recordings show that for music, this algorithm outperforms current source separation schemes.
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.