We propose a regularization method, so-called temporal smoothness regularization, for training deep neural networks for automatic sleep staging in small data settings. In intuition, we constrain the cross-entropy losses of any two adjacent epochs in the sequential input to be as close to each other as possible. The regularization closely reflects the slow transition nature of sleep process which implies small information changes between two consecutive sleep epochs. Via the regularization, we essentially discourage the network from overfitting to these small changes. Our experiments show that training the SeqSleepNet base network with the proposed regularization leads to performance improvement over the baseline without the regularization applied. Furthermore, our developed method achieves the performance on par with the state-of-the-art performance while outperforming other existing methods.

Personalized longitudinal disease assessment is central to quickly diagnosing, appropriately managing, and optimally adapting the therapeutic strategy of multiple sclerosis (MS). It is also important for identifying idiosyncratic subject-specific disease profiles. Here, we design a novel longitudinal model to map individual disease trajectories in an automated way using smartphone sensor data that may contain missing values. First, we collect digital measurements related to gait and balance, and upper extremity functions using sensor-based assessments administered on a smartphone. Next, we treat missing data via imputation. We then discover potential markers of MS by employing a generalized estimation equation. Subsequently, parameters learned from multiple training datasets are ensembled to form a simple, unified longitudinal predictive model to forecast MS over time in previously unseen people with MS. To mitigate potential underestimation for individuals with severe disease scores, the final model incorporates additional subject-specific fine-tuning using data from the first day. The results show that the proposed model is promising to achieve personalized longitudinal MS assessment; they also suggest that features related to gait and balance as well as upper extremity function, remotely collected from sensor-based assessments, may be useful digital markers for predicting MS over time.

Human sleep is cyclical with a period of approximately 90 minutes, implying long temporal dependency in the sleep data. Yet, exploring this long-term dependency when developing sleep staging models has remained untouched. In this work, we show that while encoding the logic of a whole sleep cycle is crucial to improve sleep staging performance, the sequential modelling approach in existing state-of-the-art deep learning models are inefficient for that purpose. We thus introduce a method for efficient long sequence modelling and propose a new deep learning model, L-SeqSleepNet, which takes into account whole-cycle sleep information for sleep staging. Evaluating L-SeqSleepNet on four distinct databases of various sizes, we demonstrate state-of-the-art performance obtained by the model over three different EEG setups, including scalp EEG in conventional Polysomnography (PSG), in-ear EEG, and around-the-ear EEG (cEEGrid), even with a single EEG channel input. Our analyses also show that L-SeqSleepNet is able to alleviate the predominance of N2 sleep (the major class in terms of classification) to bring down errors in other sleep stages. Moreover the network becomes much more robust, meaning that for all subjects where the baseline method had exceptionally poor performance, their performance are improved significantly. Finally, the computation time only grows at a sub-linear rate when the sequence length increases.

Automating sleep staging is vital to scale up sleep assessment and diagnosis to serve millions experiencing sleep deprivation and disorders and enable longitudinal sleep monitoring in home environments. Learning from raw polysomnography signals and their derived time-frequency image representations has been prevalent. However, learning from multi-view inputs (e.g., both the raw signals and the time-frequency images) for sleep staging is difficult and not well understood. This work proposes a sequence-to-sequence sleep staging model, XSleepNet,1that is capable of learning a joint representation from both raw signals and time-frequency images. Since different views may generalize or overfit at different rates, the proposed network is trained such that the learning pace on each view is adapted based on their generalization/overfitting behavior. In simple terms, the learning on a particular view is speeded up when it is generalizing well and slowed down when it is overfitting. View-specific generalization/overfitting measures are computed on-the-fly during the training course and used to derive weights to blend the gradients from different views. As a result, the network is able to retain the representation power of different views in the joint features which represent the underlying distribution better than those learned by each individual view alone. Furthermore, the XSleepNet architecture is principally designed to gain robustness to the amount of training data and to increase the complementarity between the input views. Experimental results on five databases of different sizes show that XSleepNet consistently outperforms the single-view baselines and the multi-view baseline with a simple fusion strategy. Finally, XSleepNet also outperforms prior sleep staging methods and improves previous state-of-the-art results on the experimental databases.

Existing generative adversarial networks (GANs) for speech enhancement solely rely on the convolution operation, which may obscure temporal dependencies across the sequence input. To remedy this issue, we propose a self-attention layer adapted from non-local attention, coupled with the convolutional and deconvolutional layers of a speech enhancement GAN (SEGAN) using raw signal input. Further, we empirically study the effect of placing the self-attention layer at the (de)convolutional layers with varying layer indices as well as at all of them when memory allows. Our experiments show that introducing self-attention to SEGAN leads to consistent improvement across the objective evaluation metrics of enhancement performance. Furthermore, applying at different (de)convolutional layers does not significantly alter performance, suggesting that it can be conveniently applied at the highest-level (de)convolutional layer with the smallest memory overhead1.

We propose in this work a multi-view learning approach for audio and music classification. Considering four typical low-level representations (i.e. different views) commonly used for audio and music recognition tasks, the proposed multi-view network consists of four subnetworks, each handling one input types. The learned embedding in the subnetworks are then concatenated to form the multi-view embedding for classification similar to a simple concatenation network. However, apart from the joint classification branch, the network also maintains four classification branches on the single-view embedding of the subnetworks. A novel method is then proposed to keep track of the learning behavior on the classification branches and adapt their weights to proportionally blend their gradients for network training. The weights are adapted in such a way that learning on a branch that is generalizing well will be encouraged whereas learning on a branch that is overfitting will be slowed down. Experiments on three different audio and music classification tasks show that the proposed multi-view network not only outperforms the single-view baselines but also is superior to the multi-view baselines based on concatenation and late fusion.

Generative adversarial networks (GAN) have recently been shown to be efficient for speech enhancement. However, most, if not all, existing speech enhancement GANs (SEGAN) make use of a single generator to perform one-stage enhancement mapping. In this work, we propose to use multiple generators that are chained to perform multi-stage enhancement mapping, which gradually refines the noisy input signals in a stage-wise fashion. Furthermore, we study two scenarios: (1) the generators share their parameters and (2) the generators' parameters are independent. The former constrains the generators to learn a common mapping that is iteratively applied at all enhancement stages and results in a small model footprint. On the contrary, the latter allows the generators to flexibly learn different enhancement mappings at different stages of the network at the cost of an increased model size. We demonstrate that the proposed multi-stage enhancement approach outperforms the one-stage SEGAN baseline, where the independent generators lead to more favorable results than the tied generators. The source code is available at http://github.com/pquochuy/idsegan.

We propose a multi-label multi-task framework based on a convolutional recurrent neural network to unify detection of isolated and overlapping audio events. The framework leverages the power of convolutional recurrent neural network architectures; convolutional layers learn effective features over which higher recurrent layers perform sequential modelling. Furthermore, the output layer is designed to handle arbitrary degrees of event overlap. At each time step in the recurrent output sequence, an output triple is dedicated to each event category of interest to jointly model event occurrence and temporal boundaries. That is, the network jointly determines whether an event of this category occurs, and when it occurs, by estimating onset and offset positions at each recurrent time step. We then introduce three sequential losses for network training: multi-label classification loss, distance estimation loss, and confidence loss. We demonstrate good generalization on two datasets: ITC-Irst for isolated audio event detection, and TUT-SED-Synthetic-2016 for overlapping audio event detection.

Acoustic scenes are rich and redundant in their content. In this work, we present a spatio-temporal attention pooling layer coupled with a convolutional recurrent neural network to learn from patterns that are discriminative while suppressing those that are irrelevant for acoustic scene classification. The convolutional layers in this network learn invariant features from time-frequency input. The bidirectional recurrent layers are then able to encode the temporal dynamics of the resulting convolutional features. Afterwards, a two-dimensional attention mask is formed via the outer product of the spatial and temporal attention vectors learned from two designated attention layers to weigh and pool the recurrent output into a final feature vector for classification. The network is trained with between-class examples generated from between-class data augmentation. Experiments demonstrate that the proposed method not only outperforms a strong convolutional neural network baseline but also sets new state-of-the-art performance on the LITIS Rouen dataset.