Musical source separation methods exploit source-specific spectral characteristics to facilitate the decomposition process. Kernel Additive Modelling (KAM) models a source applying robust statistics to time-frequency bins as specified by a source-specific kernel, a function defining similarity between bins. Kernels in existing approaches are typically defined using metrics between single time frames. In the presence of noise and other sound sources information from a single-frame, however, turns out to be unreliable and often incorrect frames are selected as similar. In this paper, we incorporate a temporal context into the kernel to provide additional information stabilizing the similarity search. Evaluated in the context of vocal separation, our simple extension led to a considerable improvement in separation quality compared to previous kernels.

Authors: Yela, Delia Fano; Ewert, Sebastian; Fitzgerald, Derry; Sandler, Mark
Affiliations: Queen Mary University of London, London, UK; Cork Institute of Technology, Cork, Ireland(See document for exact affiliation information.)
AES Conference: 2017 AES International Conference on Semantic Audio (June 2017)
Paper Number: 1-2
Publication Date: June 13, 2017 Import into BibTeX
Subject: Audio Source Separation
Permalink: https://www.aes.org/e-lib/browse.cfm?elib=18752

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.

Learn more about the AES E-Library

E-Library Location:   /conf/2017/semantic/semantic_audio_2017_paper_32.pdf

Start a discussion about this paper!