Recently, Kernel Additive Modelling was proposed as a new framework for performing sound source separation. Kernel Additive Modelling assumes that a source at some location can be estimated using its values at nearby locations where nearness is defined through a source-specific proximity kernel. Different proximity kernels can be used for different sources, which are then separated using an iterative kernel backfitting algorithm. These kernels can efficiently account for features such as continuity, stability in time or frequency and self-similarity. Here, we show that Kernel Additive Modelling can be used to generalise, extend and improve on a widely-used harmonic/percussive separation algorithm which attempts to separate pitched from percussive instruments.

International audienceIn this paper, we report the results of the 2016 community-based Signal Separation Evaluation Campaign (SiSEC 2016). This edition comprises four tasks. Three focus on the separation of speech and music audio recordings, while one concerns biomedical signals. We summarize these tasks and the performance of the submitted systems, as well as provide a small discussion concerning future trends of SiSEC

International audienceSource separation consists of separating a signal into additive components. It is a topic of considerable interest with many applications that has gathered much attention recently. Here, we introduce a new framework for source separation called Kernel Additive Modelling, which is based on local regression and permits efficient separation of multidimensional and/or nonnegative and/or non-regularly sampled signals. The main idea of the method is to assume that a source at some location can be estimated using its values at other locations nearby, where nearness is defined through a source-specific proximity kernel. Such a kernel provides an efficient way to account for features like periodicity, continuity, smoothness, stability over time or frequency, self-similarity, etc. In many cases, such local dynamics are indeed much more natural to assess than any global model such as a tensor factorization. This framework permits one to use different proximity kernels for different sources and to separate them using the iterative kernel backfitting algorithm we describe. As we show, kernel additive modelling generalizes many recent and efficient techniques for source separation and opens the path to creating and combining source models in a principled way. Experimental results on the separation of synthetic and audio signals demonstrate the effectiveness of the approach

No abstract