SDJun 15, 2017

Investigating the Potential of Pseudo Quadrature Mirror Filter-Banks in Music Source Separation Tasks

arXiv:1706.04924v1
Originality Synthesis-oriented
AI Analysis

This work addresses the challenge of separating musical sources from single-channel mixtures, which is incremental as it explores an alternative representation rather than introducing a new paradigm.

The paper tackled the problem of music source separation by investigating the use of an optimized pseudo quadrature mirror filter-bank (PQMF) as a time-frequency representation, finding that it maintains desirable sparsity and disjointness properties for this task.

Estimating audio and musical signals from single channel mixtures often, if not always, involves a transformation of the mixture signal to the time-frequency (T-F) domain in which a masking operation takes place. Masking is realized as an element-wise multiplication of the mixture signal's T-F representation with a ratio of computed sources' spectrogram. Studies have shown that the performance of the overall source estimation scheme is subject to the sparsity and disjointness properties of a given T-F representation. In this work we investigate the potential of an optimized pseudo quadrature mirror filter-bank (PQMF), as a T-F representation for music source separation tasks. Experimental results, suggest that the PQMF maintains the aforementioned desirable properties and can be regarded as an alternative for representing mixtures of musical signals.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes