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

A generative adversarial network (GAN)-based vocoder trained with an adversarial discriminator is commonly used for speech synthesis because of its fast, lightweight, and high-quality characteristics. However, this data-driven model requires a large amount of training data incurring high data-collection costs. To address this issue, we propose an augmentation-conditional discriminator (AugCondD) that receives the augmentation state as input in addition to speech, thereby assessing input speech according to augmentation state, without inhibiting the learning of the original non-augmented distribution. Experimental results indicate that AugCondD improves speech quality under limited data conditions while achieving comparable speech quality under sufficient data conditions.

In this paper, we present our submission to the 2nd e-Prevention Grand Challenge hosted at ICASSP 2024. The objective posed in the challenge was to identify psychotic and non- psychotic relapses in patients using biosignals captured by wearable sensors. Our proposed solution is an unsupervised anomaly detection approach based on Transformers. We train individual models for each patient to predict the timestamps of biosignal measurements on non-relapse days, implicitly modeling normal daily routines.

Recurrently refining the optical flow based on a single highresolution feature demonstrates high performance. We exploit the strength of this strategy to build a novel architecture for the joint learning of optical flow and depth. Our proposed architecture is improved to work in the case of training on unlabeled data, which is extremely challenging. The loss is computed for the iterations carried out over a single high-resolution feature, where the reconstruction loss fails to optimize the accuracy particularity in occluded regions.

Traditional Time-series Anomaly Detection (TAD) methods often struggle with the composite nature of complex time-series data and a diverse array of anomalies. We introduce TADNet, an end-to-end TAD model that leverages Seasonal-Trend Decomposition to link various types of anomalies to specific decomposition components, thereby simplifying the analysis of complex time-series and enhancing detection performance. Our training methodology, which includes pre-training on a synthetic dataset followed by fine-tuning, strikes a balance between effective decomposition and precise anomaly detection.

Previous Multimodal Information based Speech Processing (MISP) challenges mainly focused on audio-visual speech recognition (AVSR) with commendable success. However, the most advanced back-end recognition systems often hit performance limits due to the complex acoustic environments. This has prompted a shift in focus towards the Audio-Visual Target Speaker Extraction (AVTSE) task for the MISP 2023 challenge in ICASSP 2024 Signal Processing Grand Challenges.

Composing simple actions into complex actions is crucial yet challenging. Existing methods largely rely on language annotations to discern composable latent semantics, which is costly and labor-intensive. In this study, we introduce a novel framework to generate compositional actions without language auxiliaries. Our approach consists of three components: Action Coupling, Conditional Action Generation, and Decoupling Refinement. Action Coupling integrates two subactions to generate pseudo-training examples.

Deploying style transfer methods on resource-constrained devices is challenging, which limits their real-world applicability. To tackle this issue, we propose using pruning techniques to accelerate various visual style transfer methods. We argue that typical pruning methods may not be well-suited for style transfer methods and present an iterative correlation-based channel pruning (ICCP) strategy for encoder-transform-decoder-based image/video style transfer models.

Self-supervised learning has been gaining attention in the field of sleep stage classification. It learns representations with unlabeled electroencephalography (EEG) signals, which alleviates the cost of labeling for specialists. However, most self-supervised approaches assume only the two augmented views from the same EEG sample is a positive pair, which suffers from the false negative problem. Therefore, we propose a new model named Temporal Augmentation and False Negative Suppression (TA-FNS) to solve the problem. Specifically, it first generates two augmented views for each EEG sample.

Background music (BGM) can enhance the video’s emotion and thus make it engaging. However, selecting an appropriate BGM often requires domain knowledge or a deep understanding of the video. This has led to the development of video-music retrieval techniques. Most existing approaches utilize pre-trained video/music feature extractors trained with different target sets to obtain average video/music-level embeddings for cross-modal matching. The drawbacks are two-fold. One is that different target sets for video/music pre-training may cause the generated embeddings difficult to match.

The generalized minimax concave (GMC) penalty is a nonconvex sparse regularizer which can preserve the overall-convexity of the sparse least squares problem. In this paper, we study the solution path of a special but important instance of the GMC model termed the scaled GMC (sGMC) model. We show that despite the nonconvexity of the regularizer, there exists a solution path of the sGMC model which is piecewise linear as a function of the regularization parameter, and we propose an efficient algorithm for computing a solution path of this type.