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

Most End-to-End (E2E) Spoken Language Understanding (SLU) networks leverage the pre-trained Automatic Speech Recognition (ASR) networks but still lack the capability to understand the semantics of utterances, crucial for the SLU task. To solve this, recently proposed studies use pre-trained Natural Language Understanding (NLU) networks. However, it is not trivial to fully utilize both pre-trained networks; many solutions were proposed, such as Knowledge Distillation (KD), cross-modal shared embedding, and network integration with Interface.


Household speaker identification with few enrollment utterances is an important yet challenging problem, especially when household members share similar voice characteristics and room acoustics. A common embedding space learned from a large number of speakers is not universally applicable for the optimal identification of every speaker in a household.


Inspired by deep learning applications in structural mechanics, we focus on how to train two predictors to model the relation between the vibrational response of a prescribed point of a wooden plate and its material properties. In particular, the eigenfrequencies of the plate are estimated via multilinear regression, whereas their amplitude is predicted by a feedforward neural network.


Graphs are fundamental mathematical structures used in various fields to model statistical and physical relationships between data, signals, and processes. In some applications, such as data processing in graphs that represent physical networks, the initial network topology is known. However, disconnections of edges in the network change the topology and may affect the signals and processes over the network. In this paper, we consider the problem of edge disconnection identification in networks by using concepts from graph signal processing.


Detecting emotions directly from a speech signal plays an important role in effective human-computer interactions. Existing speech emotion recognition models require massive computational and storage resources, making them hard to implement concurrently with other machine-interactive tasks in embedded systems. In this paper, we propose an efficient and lightweight fully convolutional neural network (FCNN) for speech emotion recognition in systems with limited hardware resources.


We propose a convolutional neural network (CNN) aided factor graphs assisted by mutual information features estimated by a neural network for seizure detection. Specifically, we use neural mutual information estimation to evaluate the correlation between different electroencephalogram (EEG) channels as features. We then use a 1D-CNN to extract extra features from the EEG signals and use both features to estimate the probability of a seizure event. Finally, learned factor graphs are employed to capture the temporal correlation in the signal.