Unsupervised Training of a Deep Clustering Model for Multichannel Blind Source Separation

Drude, Lukas; Hasenklever, Daniel; Haeb‐Umbach, Reinhold

doi:10.1109/icassp.2019.8683520

Cited by 42 publications

(38 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike the proposed method, CACGMM does not have a reverberation model. Originally, a CACGMM based method without phase difference feature was proposed in [19]. However, the proposed method utilizes phase difference between microphones as an input feature.…”

Section: Resultsmentioning

confidence: 99%

“…For speech source separation performance evaluation, we utilized SDR and SIR from BSS EVAL [29]. Four methods were evaluated, i.e., 1) Conventional unsupervised training method with complex angular central Gaussian mixture model (CACGMM) [19]: Time-frequency mask of each source is inferred with the sparseness assumption. This model does not have any reverberation model.…”

Section: Evaluation Measures and Comparative Methodsmentioning

confidence: 99%

“…Recently, unsupervised DNN training techniques have been proposed [19,20]. These techniques estimate a time-frequency mask based on the DC.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Unsupervised Training for Deep Speech Source Separation with Kullback-Leibler Divergence Based Probabilistic Loss Function

Togami

Masuyama

Komatsu

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we propose a multi-channel speech source separation with a deep neural network (DNN) which is trained under the condition that no clean signal is available. As an alternative to a clean signal, the proposed method adopts an estimated speech signal by an unsupervised speech source separation with a statistical model. As a statistical model of microphone input signal, we adopts a timevarying spatial covariance matrix (SCM) model which includes reverberation and background noise submodels so as to achieve robustness against reverberation and background noise. The DNN infers intermediate variables which are needed for constructing the timevarying SCM. Speech source separation is performed in a probabilistic manner so as to avoid overfitting to separation error. Since there are multiple intermediate variables, a loss function which evaluates a single intermediate variable is not applicable. Instead, the proposed method adopts a loss function which evaluates the output probabilistic signal directly based on Kullback-Leibler Divergence (KLD). Gradient of the loss function can be back-propagated into the DNN through all the intermediate variables. Experimental results under reverberant conditions show that the proposed method can train the DNN efficiently even when the number of training utterances is small, i.e., 1K.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Evaluation Measures and Comparative Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Training for Deep Speech Source Separation with Kullback-Leibler Divergence Based Probabilistic Loss Function

Togami

Masuyama

Komatsu

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Unsupervised training for neural source separation using multichannel mixture signals has recently gained a lot of attention [12][13][14][15]. One approach is to generate supervised data by using multichannel separation methods [12][13][14]. This approach suffers from the estimation errors of the multichannel methods mentioned above.…”

Section: Introductionmentioning

confidence: 99%

Deep Bayesian Unsupervised Source Separation Based On A Complex Gaussian Mixture Model

Bando

Sasaki

Yoshii

2019

2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)

View full text Add to dashboard Cite

This paper presents an unsupervised method that trains neural source separation by using only multichannel mixture signals. Conventional neural separation methods require a lot of supervised data to achieve excellent performance. Although multichannel methods based on spatial information can work without such training data, they are often sensitive to parameter initialization and degraded with the sources located close to each other. The proposed method uses a cost function based on a spatial model called a complex Gaussian mixture model (cGMM). This model has the time-frequency (TF) masks and direction of arrivals (DoAs) of sources as latent variables and is used for training separation and localization networks that respectively estimate these variables. This joint training solves the frequency permutation ambiguity of the spatial model in a unified deep Bayesian framework. In addition, the pre-trained network can be used not only for conducting monaural separation but also for efficiently initializing a multichannel separation algorithm. Experimental results with simulated speech mixtures showed that our method outperformed a conventional initialization method.

show abstract

“…Consequently, supervised source separation using neural networks relies on the availability of paired mixtureclean data in the training set and cannot be used when such paired datasets are not available or expensive to collect. To relax these constraints, a few recent papers use other forms of information like the spatial separation between the sources in a multi-microphone setting, to train the networks for unsupervised source separation [7,8,9,10]. However, these constraints continue to impose restrictions on singlechannel source separation, where such secondary forms of information about the sources are not available.…”

Section: Introductionmentioning

confidence: 99%

A Style Transfer Approach to Source Separation

Venkataramani

Tzinis

Smaragdis

2019

2019 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)

View full text Add to dashboard Cite

Training neural networks for source separation involves presenting a mixture recording at the input of the network and updating network parameters in order to produce an output that resembles the clean source. Consequently, supervised source separation depends on the availability of paired mixture-clean training examples. In this paper, we interpret source separation as a style transfer problem. We present a variational auto-encoder network that exploits the commonality across the domain of mixtures and the domain of clean sounds and learns a shared latent representation across the two domains. Using these cycle-consistent variational auto-encoders, we learn a mapping from the mixture domain to the domain of clean sounds and perform source separation without explicitly supervising with paired training examples.

show abstract

Unsupervised Training of a Deep Clustering Model for Multichannel Blind Source Separation

Cited by 42 publications

References 32 publications

Unsupervised Training for Deep Speech Source Separation with Kullback-Leibler Divergence Based Probabilistic Loss Function

Unsupervised Training for Deep Speech Source Separation with Kullback-Leibler Divergence Based Probabilistic Loss Function

Deep Bayesian Unsupervised Source Separation Based On A Complex Gaussian Mixture Model

A Style Transfer Approach to Source Separation

Contact Info

Product

Resources

About