In maritime trajectory analysis, understanding the significance of specific locations visited by vessels is essential for navigation management, infrastructure planning, and situational awareness. This work focuses on defining, identifying, contextualizing, and classifying Zones of Interest (ZOIs), i.e., spatial areas that reflect meaningful places in vessel trajectories such as ports, anchorages, or transit slowdowns. Each ZOI is characterized by a geographic polygon, a semantic type (e.g., terminal, anchorage), and additional metadata.

We present a head-related transfer function (HRTF) estimation method which relies on a data-driven prior given by a score-based diffusion model. The HRTF is estimated in reverberant environments using natural excitation signals, e.g. human speech. The impulse response of the room is estimated along with the HRTF by optimizing a parametric model of reverberation based on the statistical behaviour of room acoustics. The posterior distribution of HRTF given the reverberant measurement and excitation signal is modelled using the score-based HRTF prior and a log-likelihood approximation. We show that the resulting method outperforms several baselines, including an oracle recommender system that assigns the optimal HRTF in our training set based on the smallest distance to the true HRTF at the given direction of arrival. In particular, we show that the diffusion prior can account for the large variability of high-frequency content in HRTFs.

Several individualization methods have recently been proposed to estimate a subject's Head-Related Transfer Function (HRTF) using convenient input modalities such as anthropometric measurements or pinnae photographs. There exists a need for adaptively correcting the estimation error committed by such methods using a few data point samples from the subject's HRTF, acquired using acoustic measurements or perceptual feedback. To facilitate this, we introduce a Convolutional Conditional Neural Process meta-learner specialized in HRTF error interpolation. In particular, the model includes a Spherical Convolutional Neural Network component to accommodate the spherical geometry of HRTF data. It also exploits potential symmetries between the HRTF's left and right channels about the median plane. In this work, we evaluate the proposed model's performance purely on time-aligned spectrum interpolation grounds under a simplified setup where a generic population-mean HRTF forms the initial estimates prior to corrections instead of individualized ones. The trained model achieves up to 3 dB relative error reduction compared to state-of-the-art interpolation methods despite being trained using only 85 subjects. This improvement translates up to nearly a halving of the data point count required to achieve comparable accuracy, in particular from 50 to 28 points to reach an average of -20 dB relative error per interpolated feature. Moreover, we show that the trained model provides well-calibrated uncertainty estimates. Accordingly, such estimates could inform the sequential decision problem of acquiring as few correcting HRTF data points as needed to meet a desired level of HRTF individualization accuracy.

Analog audio effects and synthesizers often owe their distinct sound to circuit nonlinearities. Faithfully modeling such significant aspect of the original sound in virtual analog software can prove challenging. The current work proposes a generic data-driven approach to virtual analog modeling and applies it to the Fender Bassman 56F-A vacuum-tube amplifier. Specifically, a feedforward variant of the WaveNet deep neural network is trained to carry out a regression on audio waveform samples from input to output of a SPICE model of the tube amplifier. The output signals are pre-emphasized to assist the model at learning the high-frequency content. The results of a listening test suggest that the proposed model accurately emulates the reference device. In particular, the model responds to user control changes, and faithfully restitutes the range of sonic characteristics found across the configurations of the original device.

The advent of mixed reality consumer products brings about a pressing need to develop and improve spatial sound rendering techniques for a broad user base. Despite a large body of prior work, the precise nature and importance of various sound localization cues and how they should be personalized for an individual user to improve localization performance is still an open research problem. Here we propose training a convolutional neural network (CNN) to classify the elevation angle of spatially rendered sounds and employing Layer-wise Relevance Propagation (LRP) on the trained CNN model. LRP provides saliency maps that can be used to identify spectral features used by the network for classification. These maps, in addition to the convolution filters learned by the CNN, are discussed in the context of listening tests reported in the literature. The proposed approach could potentially provide an avenue for future studies on modeling and personalization of head-related transfer functions (HRTFs).

With the rapid growth of cell phone networks during the last decades, call detail records (CDR) have been used as approximate indicators for large scale studies on human and urban mobility. Although coarse and limited, CDR are a real marker of human presence. In this paper, we use more than 800 million CDR to identify weekly patterns of human mobility through mobile phone data. Our methodology is based on the classification of individuals into six distinct presence profiles where we focus on the inherent temporal and geographical characteristics of each profile within a territory. Then, we use an event-based algorithm to cluster individuals and we identify 12 weekly patterns. We leverage these results to analyze population estimates adjustment processes and as a result, we propose new indicators to characterize the dynamics of a territory. Our model has been applied to real data coming from more than 1.6 million individuals and demonstrates its relevance. The product of our work can be used by local authorities for human mobility analysis and urban planning.

Spatial and temporal decomposition of aggregated mobility flows is nowadays a commonly addressed issue, but a trip-purpose decomposition of mobility flows is a more challenging topic, which requires more sensitive analysis such as heterogeneous data fusion. In this paper, we study the relation between land use and mobility purposes. We propose a model that dynamically decomposes mobility flows into six mobility purposes. To this end, we use a national transportation database that surveyed more than 35,000 individuals and a national ground description database that identifies six distinct ground types. Based on these two types of data, we dynamically solve several overdetermined systems of linear equations from a training set and we infer the travel purposes. Our experimental results demonstrate that our model effectively predicts the purposes of mobility from the land use. Furthermore, our model shows great results compared with a reference supervised learning decomposition.