Blind One-microphone Speech Separation: A Spectral Learning Approach

116 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.

Music Source Separation in the Waveform Domain (2019, methodology): “Learning a soft/binary mask over power spectrograms has been done using either HMM-based prediction [Roweis, 2001] or segmentation techniques [Bach and Jordan, 2005].”
A Supervised Learning Approach to Monaural Segregation of Reverberant Speech (2009, background): “Studies show that speech reconstructed from the idea binary mask produces large improvement in human speech intelligibility [3,12,32].”
Single-channel speech separation using sparse non-negative matrix factorization (2006, background): “A number researchers have taken ideas from the computational auditory scene analysis (CASA) literature, trying to incorpo- rate various grouping cues of the human auditory system in speech separation algorithms [9, 10].”
Soft Mask Methods for Single-Channel Speaker Separation (2007, background): “Bach and Jordan [12] take a more data-driven approach of representing time–frequency components by perceptually motivated features and grouping them by a procedure referred to as spectral clustering (where the term “spectrum” refers to the eigen spectrum and is not to be confused with the frequency ”