Sorry, you need to enable JavaScript to visit this website.

In this work, we present a language identification (LID) system based on embeddings. In our case, an embedding is a fixed-length vector (similar to i-vector) that represents the whole utterance, but unlike i-vector it is designed to contain mostly information relevant to the target task (LID). In order to obtain these embeddings, we train a deep neural network (DNN) with sequence summarization layer to classify languages.

Artificial bandwidth extension (ABE) algorithms have been developed to improve speech quality when wideband devices are used in conjunction with narrowband devices or infrastructure. While past work points to the benefit of using contextual information or memory for ABE, an understanding of the relative benefit of explicit memory inclusion, rather than just dynamic information, calls for a comparative, quantitative analysis. The need for practical ABE solutions calls further for the inclusion of memory without significant increases to latency or computational complexity.

We investigate the practical realization of energy beamforming gains in the downlink wireless power transfer from a massive antenna radio frequency (RF) source to multiple single antenna energy harvesting (EH) users. Assuming channel reciprocity for the uplink and downlink channels undergoing Rician fading, we first obtain the least-squares and linear-minimum-mean-square-error channel estimates using the energy-constrained pilot signal transmission from EH users.

Millimeter wave (mmWave) multiple-input multiple-output (MIMO) transceivers employ narrow beams to obtain a large array-gain, rendering them sensitive to changes in the angles of arrival and departure of the paths. Since the singular vectors that span the channel subspace are used to design the precoder and combiner, we propose a method to track the receiver-side channel subspace during data transmission using a separate radio frequency (RF) chain dedicated for channel tracking.

Random sample consensus (RANSAC) is a popular paradigm for parameter estimation with outlier detection, which plays an essential role in 3D robot vision, especially for LiDAR odometry. The success of RANSAC strongly depends on the probability of selecting a subset of pure inliers, which sets barriers to robust and fast parameter estimation. Although significant efforts have been made to improve RANSAC in various scenarios, its strong dependency on inlier selection is still a problem.

This paper revisits the Degenerate Unmixing Estimation Technique (DUET) for blind audio separation of an arbitrary
number of sources given two mixtures through a recursively computed and adaptive time-frequency representation.
Recently, synchrosqueezing was introduced as a promising signal disentangling method which allows to compute reversible
and sharpen time-frequency representations. Thus, it can be used to reduce overlaps between the sources in the

The paper provides an analysis of automatic speech recognition
systems (ASR) based on multilingual BLSTM, where we used multi-task
training with separate classification layer for each language. The
focus is on low resource languages, where only a limited
amount of transcribed speech is available. In such
scenario, we found it
essential to train the ASR systems in a multilingual fashion and we
report superior results
obtained with pre-trained multilingual BLSTM on this task.
The high resource languages are also