Underdetermined Convolutive Blind Source Separation via Frequency Bin-Wise Clustering and Permutation Alignment

By Hiroshi Sawada, Shoko Araki, Shoji Makino. IEEE Transactions on Audio, Speech, and Language Processing, 2011

370 citations recorded by the source corpus.

Subfields: Computational Mechanics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/5c0c9324d551cac7c553f65c11be0f9f2528ffc7

How researchers see this paper

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

In Computational Mechanics

In Signal Processing

Citations in the corpus

A Consolidated Perspective on Multimicrophone Speech Enhancement and Source Separation (2017, neutral): “In [47], the inter-frame model and the full-rank model are combined in a nested MVDR beamforming structure.”
A Review of Speaker Diarization: Recent Advances with Deep Learning (2021, methodology): “speech separation methods have been proposed based on Independent Component Analysis(ICA) or Spatial Clustering [210, 211, 212, 213, 214, 215], but applying them to a meeting setup had limited success.”
The NTT CHiME-3 system: Advances in speech enhancement and recognition for mobile multi-microphone devices (2015, methodology): “Many spectral mask estimation schemes have been proposed by the speech separation community, including those based on GMMs [30, 31], Watson mixture models (WMMs) [32, 33] and complex GMMs (CG-MMs) 1 [34].”
Global and Local Simplex Representations for Multichannel Source Separation (2020, methodology): “Severalmethodsexistforsolvingthispermutation ambiguityproblem,forinstance,basedontheTDOAorbasedon similartime-activitypatternsofneighbouringfrequencies[18], [21].”
Speaker-Aware Neural Network Based Beamformer for Speaker Extraction in Speech Mixtures (2017, methodology): “This problem has been investigated using both single and multi-channel source separation techniques, including NMF [1], ICA [2] or clustering of spatial cues [3, 4, 5].”
LINEAR PREDICTION-BASED DEREVERBERATION WITH ADVANCED SPEECH ENHANCEMENT AND RECOGNITION TECHNOLOGIES FOR THE REVERB CHALLENGE (2014, methodology): “DOLPHIN efficiently combines conventional direction-of-arrival (DOA) [19, 20] feature based enhancement [21] and spectral feature based approaches through a dominant source index (DSI) that indicates whether noise or speech is dominant at each time/frequency bin.”
Cracking the cocktail party problem by multi-beam deep attractor network (2017, methodology): “posed by using, for example, Independent Component Analysis [16, 17]. Alternatively, clustering-based algorithms attempt to cluster time-frequency bins to individual sources by using spatial features [18, 19, 20]. There is also a clustering-based approach that uses both spatial and spectral features”
Direction of Arrival Based Spatial Covariance Model for Blind Sound Source Separation (2014, neutral): “Separation methods directly utilizing TDoAs between microphones and creating time-frequency separation masks by clustering the measured TDoAs at each frequency include for example DUET [10] and binwise clustering [11].”
Neural network based spectral mask estimation for acoustic beamforming (2016, background): “[3, 4, 5, 6, 7, 8]), we want to leverage the power of a discriminatively trained data-driven approach to estimate a spectral mask for the speech and the noise component.”