Filters








2,694 Hits in 4.5 sec

Direction of Arrival Based Spatial Covariance Model for Blind Sound Source Separation

Joonas Nikunen, Tuomas Virtanen
2014 IEEE/ACM Transactions on Audio Speech and Language Processing  
Index Terms-multichannel source separation, spatial covariance models, non-negative matrix factorization, direction of arrival estimation, array signal processing  ...  This paper addresses the problem of sound source separation from a multichannel microphone array capture via estimation of source spatial covariance matrix (SCM) of a shorttime Fourier transformed mixture  ...  CONCLUSION In this paper we have proposed a direction of arrival (DoA) based spatial covariance matrix (SCM) model for the purpose of spatial sound source separation using complex-valued nonnegative matrix  ... 
doi:10.1109/taslp.2014.2303576 fatcat:5mzcsuhtvbbr5odard25xxmjke

Under-determined convolutive blind source separation using spatial covariance models

Ngoc Q.K. Duong, Emmanuel Vincent, Remi Gribonval
2010 2010 IEEE International Conference on Acoustics, Speech and Signal Processing  
We then align the order of the estimated sources across all frequency bins based on their estimated directions of arrival (DOA).  ...  This paper deals with the problem of under-determined convolutive blind source separation.  ...  INTRODUCTION In blind source separation (BSS), the recorded multichannel signal is a mixture of several sound sources (1) where is the spatial image of source , that is its contribution to all mixture  ... 
doi:10.1109/icassp.2010.5496284 dblp:conf/icassp/DuongVG10 fatcat:kg6yktbndnhoths5kepwo7diom

Under-Determined Reverberant Audio Source Separation Using a Full-Rank Spatial Covariance Model

Ngoc Q K Duong, Emmanuel Vincent, Rémi Gribonval
2010 IEEE Transactions on Audio, Speech, and Language Processing  
This paper addresses the modeling of reverberant recording environments in the context of under-determined convolutive blind source separation.  ...  We model the contribution of each source to all mixture channels in the time-frequency domain as a zero-mean Gaussian random variable whose covariance encodes the spatial characteristics of the source.  ...  Izumi for providing the results of his algorithm [15] on some test data.  ... 
doi:10.1109/tasl.2010.2050716 fatcat:rvzamjlorfdppj4u7lghrmenxm

Under-determined reverberant audio source separation using a full-rank spatial covariance model [article]

Ngoc Duong, Remi Gribonval
2009 arXiv   pre-print
estimated sources across all frequency bins based on their estimated directions of arrival (DOA).  ...  This article addresses the modeling of reverberant recording environments in the context of under-determined convolutive blind source separation.  ...  INRIA Conclusion and discussion In this article, we presented a general probabilistic framework for convolutive source separation based on the notion of spatial covariance matrix.  ... 
arXiv:0912.0171v2 fatcat:tidimxfzavfn7echpfz5nlqwri

Multichannel audio separation by direction of arrival based spatial covariance model and non-negative matrix factorization

Joonas Nikunen, Tuomas Virtanen
2014 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
The proposed model for SCMs is parameterized by source direction of arrival (DoA) and its parameters can be optimized to yield a spatially coherent solution over frequencies thus avoiding permutation ambiguity  ...  This paper studies multichannel audio separation using non-negative matrix factorization (NMF) combined with a new model for spatial covariance matrices (SCM).  ...  In this paper, we introduce a direction of arrival (DoA) based SCM model for spatial audio separation and use NMF as the source magnitude model.  ... 
doi:10.1109/icassp.2014.6854892 dblp:conf/icassp/NikunenV14 fatcat:c3nzyo2it5e4vi573eukqnaxxu

Blind Subband Beamforming With Time-Delay Constraints for Moving Source Speech Enhancement

Zohra Yermeche, Nedelko Grbic, Ingvar Claesson
2007 IEEE Transactions on Audio, Speech, and Language Processing  
The novelty of the proposed method involves considering the spatial spreading of the sound source as equivalent to a time-delay spreading, thus, allowing for the estimated intersensor time-delays to be  ...  This blind approach is based on a two-stage scheme. First, a subband time-delay estimation method is used to localize the dominant speech source.  ...  Hence, using the general model for a spatially spread source given in (8) , the source covariance matrix can be estimated from the direct-path signal as (9) where is the source power spectral density  ... 
doi:10.1109/tasl.2007.903309 fatcat:rlzdq3vac5edjjgsy6dygh3ctq

Real‐time acoustic blind signal separation system based on the spatio‐temporal gradient analysis

Kenbu Teramoto, Md. Tawhidul Islam Khan
2008 Journal of the Acoustical Society of America  
Acoustics 08 Paris 111 This paper presents an autonomous directivity microphone system for the blind source separation based on the newly proposed spatio-temporal gradient algorithm.  ...  Several acoustical experiments have been performed with acceptable performance of the proposed method for the real-time acoustic blind source separation. where, c denotes the phase velocity of the airborne  ...  For detecting the directions of arrivals (DOA), the MS-microphone system has been used as the (2) R-ch. analysis shows the linear dependency among source sound pressures.  ... 
doi:10.1121/1.2932891 fatcat:rr5bvfdt3vddzoqce6t3i54pja

Separation of Moving Sound Sources Using Multichannel NMF and Acoustic Tracking [article]

Joonas Nikunen, Aleksandr Diment, Tuomas Virtanen
2017 arXiv   pre-print
The SCMs of the model are obtained based on estimated directions of arrival of tracked sources at each time frame.  ...  In this paper we propose a method for separation of moving sound sources.  ...  Alternatively, separation of moving sources can be achieved by tracking the spatial position or direction of arrival (DOA) of the sources and using spatial filtering (beamforming or separation mask) for  ... 
arXiv:1710.10005v1 fatcat:wi5t6xm3s5affh7aayvxsfwnb4

2020 Index IEEE/ACM Transactions on Audio, Speech, and Language Processing Vol. 28

2020 IEEE/ACM Transactions on Audio Speech and Language Processing  
., +, TASLP 2020 1129-1142 Fast Multichannel Nonnegative Matrix Factorization With Directivity-Aware Jointly-Diagonalizable Spatial Covariance Matrices for Blind Source Separation.  ...  Boulianne, G., TASLP 2020 2781-2795 Fast Multichannel Nonnegative Matrix Factorization With Directivity-Aware Jointly-Diagonalizable Spatial Covariance Matrices for Blind Source Separation.  ...  T Target tracking Multi-Hypothesis Square-Root Cubature Kalman Particle Filter for Speaker Tracking in Noisy and Reverberant Environments. Zhang, Q., +, TASLP 2020 1183 -1197  ... 
doi:10.1109/taslp.2021.3055391 fatcat:7vmstynfqvaprgz6qy3ekinkt4

Spatial location priors for Gaussian model based reverberant audio source separation

Ngoc Q K Duong, Emmanuel Vincent, Rémi Gribonval
2013 EURASIP Journal on Advances in Signal Processing  
We consider the Gaussian framework for reverberant audio source separation, where the sources are modeled in the time-frequency domain by their short-term power spectra and their spatial covariance matrices  ...  A priori de localisation spatiale pour la séparation de sources audio réverbérées par modèle gaussien Résumé : Nous nous plaçons dans le cadre gaussien pour la séparation de mélanges réverbérants de sources  ...  Fixed spatial covariance matrices set to the value in (7) were employed for single source localization in [29] and for source separation in [30] .  ... 
doi:10.1186/1687-6180-2013-149 fatcat:loj6jszytvhm3jccca4ig4a7sm

Adaptive blind source separation with HRTFs beamforming preprocessing

Mounira Maazaoui, Karim Abed-Meraim, Yves Grenier
2012 2012 IEEE 7th Sensor Array and Multichannel Signal Processing Workshop (SAM)  
We propose an adaptive blind source separation algorithm in the context of robot audition using a microphone array.  ...  In the source separation step, we use a separation algorithm based on the l1 norm minimization.  ...  a rough estimation of the directions of arrival (DOA).  ... 
doi:10.1109/sam.2012.6250486 dblp:conf/ieeesam/MaazaouiAG12 fatcat:kfd6hvk3y5fwfacvnresghraim

Multi-Channel Overlapped Speech Recognition with Location Guided Speech Extraction Network

Zhuo Chen, Xiong Xiao, Takuya Yoshioka, Hakan Erdogan, Jinyu Li, Yifan Gong
2018 2018 IEEE Spoken Language Technology Workshop (SLT)  
In the proposed system, three different features are formed for each target speaker, namely, spectral, spatial, and angle features.  ...  To solve these problems, beamforming and speech separation networks were previously proposed. However, they tend to suffer from leakage of interfering speech or limited generalizability.  ...  To get the angle feature, we first form the steering vector for the direction-of-arrival (DOA) of each speaker.  ... 
doi:10.1109/slt.2018.8639593 dblp:conf/slt/ChenXYELG18 fatcat:lwfz7dkatzejhmc72uhxj4aufy

GEV Beamforming Supported by DOA-based Masks Generated on Pairs of Microphones [article]

Francois Grondin, Jean-Samuel Lauzon, Jonathan Vincent, Francois Michaud
2020 arXiv   pre-print
Recently, a combination of beamforming and speech separation networks have been proposed to improve the target source quality in the direction of arrival of interest.  ...  Sound source separation is thus often required as a preprocessing step prior to speech recognition to improve the signal to distortion ratio (SDR).  ...  Blind speech separation relies strictly on the mixture spectrogram to restore the individual sources, whereas informed speech separation uses additional information such as video, direction of arrival  ... 
arXiv:2005.09587v2 fatcat:66ko3i6edzcpzazquhi3a5zhgi

GEV Beamforming Supported by DOA-Based Masks Generated on Pairs of Microphones

François Grondin, Jean-Samuel Lauzon, Jonathan Vincent, François Michaud
2020 Interspeech 2020  
Recently, a combination of beamforming and speech separation networks have been proposed to improve the target source quality in the direction of arrival of interest.  ...  Sound source separation is thus often required as a preprocessing step prior to speech recognition to improve the signal to distortion ratio (SDR).  ...  Blind speech separation relies strictly on the mixture spectrogram to restore the individual sources, whereas informed speech separation uses additional information such as video, direction of arrival  ... 
doi:10.21437/interspeech.2020-2687 dblp:conf/interspeech/GrondinLVM20 fatcat:ducuupqxgrfsldp27ujwystkdy

From Blind to Guided Audio Source Separation: How models and side information can improve the separation of sound

Emmanuel Vincent, Nancy Bertin, Remi Gribonval, Frederic Bimbot
2014 IEEE Signal Processing Magazine  
After a brief historical account, we provide an overview of recent and ongoing research in this field, illustrating a variety of models and techniques designed so as to guide the audio source separation  ...  Starting with blind separation of toy mixtures in the mid 90's, research has progressed up to real-world scenarios today, with applications to speech enhancement and recognition, music editing, 3D sound  ...  While early source separation techniques relied on spatial diversity, that is the assumption that the sources have different directions of arrival, the move to time-frequency domain processing enabled  ... 
doi:10.1109/msp.2013.2297440 fatcat:klya2wqk45ecrcy4t7ssz3xcne
« Previous Showing results 1 — 15 out of 2,694 results