Separation of Mixed Audio Sources By Independent Subspace Analysis

By Michael A. Casey, Alex Westner. International Conference on Mathematics and Computing, 2000

316 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Convolutive Speech Bases and Their Application to Supervised Speech Separation (2007, background): “Results relating this approach to source separation have been reported in multiple publications [1], [2], [12], [15] and have been a promising field of research for some time.”
Source separation using single channel ICA (2007, methodology): “In [7] grouping was done based on cross entropy estimates between components while in [8] synchronization of the analysis blocks with the primary source was used.”
Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style (2021, background): “1 is related to independent subspace analysis [16, 53, 73, 114], which also aims to identify blocks of random variables as opposed to individual factors, though under an independence assumption across blocks, and typically not within a multi-view setting as studied in the present work.”
Monaural Sound Source Separation by Nonnegative Matrix Factorization With Temporal Continuity and Sparseness Criteria (2007, methodology): “Implementation of the algorithm follows the outline proposed by Casey and Westner [5] , but instead of aiming at statistically independent gains, we estimated statistically independent spectra, since that produced better results.”