Model-based Source Separation Research Articles

A model-based source separation algorithm for lung perfusion imaging using electrical impedance tomography

Objective. Electrical impedance tomography (EIT) for lung perfusion imaging is attracting considerable interest in intensive care, as it might open up entirely new ways to adjust ventilation therapy. A promising technique is bolus injection of a conductive indicator to the central venous catheter, which yields the indicator-based signal (IBS). Lung perfusion images are then typically obtained from the IBS using the maximum slope technique. However, the low spatial resolution of EIT results in a partial volume effect (PVE), which requires further processing to avoid regional bias. Approach. In this work, we repose the extraction of lung perfusion images from the IBS as a source separation problem to account for the PVE. We then propose a model-based algorithm, called gamma decomposition (GD), to derive an efficient solution. The GD algorithm uses a signal model to transform the IBS into a parameter space where the source signals of heart and lung are separable by clustering in space and time. Subsequently, it reconstructs lung model signals from which lung perfusion images are unambiguously extracted. Main results. We evaluate the GD algorithm on EIT data of a prospective animal trial with eight pigs. The results show that it enables lung perfusion imaging using EIT at different stages of regional impairment. Furthermore, parameters of the source signals seem to represent physiological properties of the cardio-pulmonary system. Significance. This work represents an important advance in IBS processing that will likely reduce bias of EIT perfusion images and thus eventually enable imaging of regional ventilation/perfusion (V/Q) ratio.

Bootstrap Averaging for Model-Based Source Separation in Reverberant Conditions

Recently proposed model-based methods use time-frequency (T-F) masking for source separation, where the T-F masks are derived from various cues described by a frequency domain Gaussian mixture model (GMM). These methods work well for separating mixtures recorded in low-to-medium level of reverberation, however, their performance degrades as the level of reverberation is increased. We note that the relatively poor performance of these methods under reverberant conditions can be attributed to the high variance of the frequency-dependent GMM parameter estimates. To address this limitation, a novel bootstrap-based approach is proposed to improve the accuracy of expectation maximization estimates of a frequency-dependent GMM based on an a priori chosen initialization scheme. It is shown how the proposed technique allows us to construct time-frequency masks which lead to improved model-based source separation for reverberant speech mixtures. Experiments and analysis are performed on speech mixtures formed using real room-recorded impulse responses.

Video-Aided Model-Based Source Separation in Real Reverberant Rooms

Source separation algorithms that utilize only audio data can perform poorly if multiple sources or reverberation are present. In this paper we therefore propose a video-aided model-based source separation algorithm for a two-channel reverberant recording in which the sources are assumed static. By exploiting cues from video, we first localize individual speech sources in the enclosure and then estimate their directions. The interaural spatial cues, the interaural phase difference and the interaural level difference, as well as the mixing vectors are probabilistically modeled. The models make use of the source direction information and are evaluated at discrete time-frequency points. The model parameters are refined with the well-known expectation-maximization (EM) algorithm. The algorithm outputs time-frequency masks that are used to reconstruct the individual sources. Simulation results show that by utilizing the visual modality the proposed algorithm can produce better time-frequency masks thereby giving improved source estimates. We provide experimental results to test the proposed algorithm in different scenarios and provide comparisons with both other audio-only and audio-visual algorithms and achieve improved performance both on synthetic and real data. We also include dereverberation based pre-processing in our algorithm in order to suppress the late reverberant components from the observed stereo mixture and further enhance the overall output of the algorithm. This advantage makes our algorithm a suitable candidate for use in under-determined highly reverberant settings where the performance of other audio-only and audio-visual methods is limited.

Speech separation using speaker-adapted eigenvoice speech models

We present a system for model-based source separation for use on single channel speech mixtures where the precise source characteristics are not known a priori. The sources are modeled using hidden Markov models (HMM) and separated using factorial HMM methods. Without prior speaker models for the sources in the mixture it is difficult to exactly resolve the individual sources because there is no way to determine which state corresponds to which source at any point in time. This is solved to a small extent by the temporal constraints provided by the Markov models, but permutations between sources remains a significant problem. We overcome this by adapting the models to match the sources in the mixture. We do this by representing the space of speaker variation with a parametric signal model-based on the eigenvoice technique for rapid speaker adaptation. We present an algorithm to infer the characteristics of the sources present in a mixture, allowing for significantly improved separation performance over that obtained using unadapted source models. The algorithm is evaluated on the task defined in the 2006 Speech Separation Challenge [Cooke, M.P., Lee, T.-W., 2008. The 2006 Speech Separation Challenge. Computer Speech and Language] and compared with separation using source-dependent models. Although performance is not as good as with speaker-dependent models, we show that the system based on model adaptation is able to generalize better to held out speakers.