No abstract

In this paper, we propose a two-step training procedure for source separation via a deep neural network. In the first step we learn a transform (and it's inverse) to a latent space where masking-based separation performance using oracles is optimal. For the second step, we train a separation module that operates on the previously learned space. In order to do so, we also make use of a scale-invariant signal to distortion ratio (SI-SDR) loss function that works in the latent space, and we prove that it lower-bounds the SI-SDR in the time domain. We run various sound separation experiments that show how this approach can obtain better performance as compared to systems that learn the transform and the separation module jointly. The proposed methodology is general enough to be applicable to a large class of neural network end-to-end separation systems.

This paper describes Asteroid, the PyTorch-based audio source separation toolkit for researchers. Inspired by the most successful neural source separation systems, it provides all neural building blocks required to build such a system. To improve reproducibility, Kaldi-style recipes on common audio source separation datasets are also provided. This paper describes the software architecture of Asteroid and its most important features. By showing experimental results obtained with Asteroid's recipes, we show that our implementations are at least on par with most results reported in reference papers. The toolkit is publicly available at github.com/mpariente/asteroid.

Deep learning approaches have recently achieved impressive performance on both audio source separation and sound classification. Most audio source separation approaches focus only on separating sources belonging to a restricted domain of source classes, such as speech and music. However, recent work has demonstrated the possibility of "universal sound separation", which aims to separate acoustic sources from an open domain, regardless of their class. In this paper, we utilize the semantic information learned by sound classifier networks trained on a vast amount of diverse sounds to improve universal sound separation. In particular, we show that semantic embeddings extracted from a sound classifier can be used to condition a separation network, providing it with useful additional information. This approach is especially useful in an iterative setup, where source estimates from an initial separation stage and their corresponding classifier-derived embeddings are fed to a second separation network. By performing a thorough hyperparameter search consisting of over a thousand experiments, we find that classifier embeddings from clean sources provide nearly one dB of SNR gain, and our best iterative models achieve a significant fraction of this oracle performance, establishing a new state-of-the-art for universal sound separation.

No abstract

We present a monophonic source separation system that is trained by only observing mixtures with no ground truth separation information. We use a deep clustering approach which trains on multichannel mixtures and learns to project spectrogram bins to source clusters that correlate with various spatial features. We show that using such a training process we can obtain separation performance that is as good as making use of ground truth separation information. Once trained, this system is capable of performing sound separation on monophonic inputs, despite having learned how to do so using multi-channel recordings.

No abstract

Genetic polymorphisms in β1-, β2- and β3-adrenergic receptors (β-ARs) have been associated with chronic non-communicable disorders, such as cardiovascular diseases, asthma, chronic obstructive pulmonary disease (COPD) and obesity, as well as β-agonists and antagonists response and toxicity. The purpose of this study was to determine the frequency distribution of ADRB1 genetic variants Ser49Gly and Arg389Gly, ADRB2 variants Gly16Arg and Gln27Glu, ADRB3 variant Trp64Arg in a Southeastern European Caucasian (SEC) population sample and to establish a comparison with existing data from other human populations. A sample of 431 men and 590 women volunteered to participate in this genotyping analysis after anonymization and de-identification. Real Time PCR (Melting Curve Analysis) followed DNA extraction from buccal swabs and statistical analysis of the results was performed. The allele frequencies in the SEC population were Ser49 (90.3%), Arg389 (69.49%), Gly16 (61.61%), Gln27 (65.72%), and Trp64 (94.52%), while a Hardy-Weinberg Equilibrium (HWE) was detected in the population studied. Comparisons for the Ser49Gly, Gln27Glu, and Trp64Arg allele distributions demonstrated significant differences between SEC and the European group. European subgroups comparisons showed that allele distributions were similar for four of the five SNPs between SEC and Southwestern European Caucasians (SWC), while they were quite distinct from the Northwestern European Caucasians (NWC). These data underline the importance of interethnic variability of β-ARs genetic polymorphisms.