Tools for Ground-Truth-Free Passive Client Density Mapping in MAC-Randomized Outdoor WiFi Networks.

In the past few years, data privacy legislation has hampered the ability of WiFi network operators to count and map client activity for commercial and security purposes. Indeed, since client device MAC devices are now randomized at each transmission, aggregating client activity using management frames such as Probe Requests, as has been common practice in the past, becomes problematic. Recently, researchers have demonstrated that, statistically, client counts are roughly proportional to raw Probe Request counts, thus somewhat alleviating the client counting problem, even if, in most cases, ground truth measurements from alternate sensors such as cameras are necessary to establish this proportionality. Nevertheless, localizing randomized MAC clients at a network site is currently an unsolved problem. In this work, we propose a set of nine tools for extending the proportionality between client counts and Probe Requests to the mapping of client densities in real-world outdoor WiFi networks without the need for ground truth measurements. The purpose of the proposed toolkit is to transform raw, randomized MAC Probe Request counts into a density map calibrated to an estimated number of clients at each position.

Future Speech Interfaces with Sensors and Machine Intelligence.

Speech is the most spontaneous and natural means of communication. Speech is also becoming the preferred modality for interacting with mobile or fixed electronic devices. However, speech interfaces have drawbacks, including a lack of user privacy; non-inclusivity for certain users; poor robustness in noisy conditions; and the difficulty of creating complex man-machine interfaces. To help address these problems, the Special Issue "Future Speech Interfaces with Sensors and Machine Intelligence" assembles eleven contributions covering multimodal and silent speech interfaces; lip reading applications; novel sensors for speech interfaces; and enhanced speech inclusivity tools for future speech interfaces. Short summaries of the articles are presented, followed by an overall evaluation. The success of this Special Issue has led to its being re-issued as "Future Speech Interfaces with Sensors and Machine Intelligence-II" with a deadline in March of 2023.

Statistical Approach to Estimating Audience from MAC-Randomized WiFi Probe Requests.

In the past few years, the ability of wireless network operators to monitor audience using control frames emitted by client devices has been compromised, both by legislation treating client MAC addresses as private information and by the difficulty of distinguishing genuine client frames from those arising from the Internet of Things or from certain enhanced services. Here, a deterministic model, based on characteristics of human activity and on seasonal trends, is used to reveal underlying client statistics in raw MAC-randomized WiFi Probe Request data. The method proposes a candidate conversion factor, X, between probe request counts and the client population, which offers plausible predictions on real-world datasets.

Convolutional neural network-based automatic classification of midsagittal tongue gestural targets using B-mode ultrasound images.

Tongue gestural target classification is of great interest to researchers in the speech production field. Recently, deep convolutional neural networks (CNN) have shown superiority to standard feature extraction techniques in a variety of domains. In this letter, both CNN-based speaker-dependent and speaker-independent tongue gestural target classification experiments are conducted to classify tongue gestures during natural speech production. The CNN-based method achieves state-of-the-art performance, even though no pre-training of the CNN (with the exception of a data augmentation preprocessing) was carried out.

A comparative study on the contour tracking algorithms in ultrasound tongue images with automatic re-initialization.

The feasibility of an automatic re-initialization of contour tracking is explored by using an image similarity-based method in the ultrasound tongue sequences. To this end, the re-initialization method was incorporated into current state-of-art tongue tracking algorithms, and a quantitative comparison was made between different algorithms by computing the mean sum of distances errors. The results demonstrate that with automatic re-initialization, the tracking error can be reduced from an average of 5-6 to about 4 pixels, a result obtained by using a large number of hand-labeled frames and similarity measurements to extract the contours, which results in improved performance.

Robust contour tracking in ultrasound tongue image sequences.

A new contour-tracking algorithm is presented for ultrasound tongue image sequences, which can follow the motion of tongue contours over long durations with good robustness. To cope with missing segments caused by noise, or by the tongue midsagittal surface being parallel to the direction of ultrasound wave propagation, active contours with a contour-similarity constraint are introduced, which can be used to provide 'prior' shape information. Also, in order to address accumulation of tracking errors over long sequences, we present an automatic re-initialization technique, based on the complex wavelet image similarity index. Experiments on synthetic data and on real 60 frame per second (fps) data from different subjects demonstrate that the proposed method gives good contour tracking for ultrasound image sequences even over durations of minutes, which can be useful in applications such as speech recognition where very long sequences must be analyzed in their entirety.

Improving ultrasound-based multimodal speech recognition with predictive features from representation learning.

Representation learning is believed to produce high-level representations of underlying dynamics in temporal sequences. A three-dimensional convolutional neural network trained to predict future frames in ultrasound tongue and optical lip images creates features for a continuous hidden Markov model based speech recognition system. Predictive tongue features are found to generate lower word error rates than those obtained from an auto-encoder without future frames, or from discrete cosine transforms. Improvement is apparent for the monophone/triphone Gaussian mixture model and deep neural network acoustic models. When tongue and lip modalities are combined, the advantage of the predictive features is reduced.

Human exposure to mercury in the vicinity of chlor-alkali plant.

The main objectives of our study were to estimate the impact of a mercury cell chlor-alkali (MCCA) complex in Rosignano Solvay (Tuscany, Italy) on the local environment and to assess mercury exposure of inhabitants living near the plant. Measurement campaigns of atmospheric Hg near the MCCA plant showed that the impact of the emitted Hg from the industry on the terrestrial environment is restricted to a close surrounding area. Total gaseous mercury concentrations in ambient air of inhabited area around the MCCA plant were in the range of 8.0-8.7 ng/m3 in summer and 2.8-4.2 ng/m3 in winter. Peaks of up to 100 ng/m3 were observed at particular meteorological conditions. Background levels of 2 ng/m3 were reached within a radius of 3 km from the plant. Reactive gaseous mercury emissions from the plant constituted around 4.2% of total gaseous mercury and total particulate mercury emission constituted around 1.0% of total gaseous mercury emitted. Analysis of local vegetables and soil samples showed relatively low concentrations of total mercury (30.1-2919 microgHg/kg DW in the soil; <0.05-111 microgHg/kg DW in vegetables) and methylmercury (0.02-3.88 microgHg/kg DW in the soil; 0.03-1.18 microgHg/kg DW in vegetables). Locally caught marine fish and fresh marine fish from the local market had concentrations of total Hg from 0.049 to 2.48 microgHg/g FW, of which 37-100% were in the form of methylmercury. 19% of analysed fish exceeded 1.0 microgHg/g FW level, which is a limit set by the European Union law on Hg concentrations in edible marine species for tuna, swordfish and shark, while 39% of analysed fish exceeded the limit of 0.5 microgHg/g FW set for all other edible marine species. Risk assessment performed by calculating ratio of probable daily intake (PDI) and provisional tolerable daily intake (PTDI) for mercury species for various exposure pathways showed no risks to human health for elemental and inorganic mercury, except for some individuals with higher number of amalgam fillings, while PDI/PTDI ratio for methylmercury and total mercury exceeded the toxicologically tolerable value due to the potential consumption of contaminated marine fish.

Urinary mercury in adults in Poland living near a chloralkali plant.

We conducted a study within the framework of the interdisciplinary European Mercury Emission from Chloralkali Plants (EMECAP) project to assess exposure to mercury (Hg) and the contribution of Hg emissions from a mercury cell chloralkali plant to urinary mercury (U-Hg) in adults living near the plant. We collected data from questionnaires and first morning urine samples from 75 subjects living near the Tarnow plant in Poland and 100 subjects living in a reference area. Median U-Hg was 0.32 mug/g creatinine (microg/gC) and 0.20 microg/gC, respectively. The median U-Hg was also higher in the amalgam-free subjects living near the plant (0.26 microg/gC) than in the reference group (0.18 microg/gC), but no such association was found in a multivariate analysis. There was a statistically significant positive association between U-Hg and number of teeth with amalgams, a negative association with age and a tendency towards higher U-Hg in female subjects. In the amalgam-free subjects there were statistically significant effects of female sex and fish consumption, and a negative association with age. The additional long-term average air Hg concentration from the plant, based on EMECAP environmental measurements and modelling, was estimated to be 1-3.5 ng/m(3) for the residential study area and should have a very small effect on U-Hg. The other Hg emission sources such as coal combustion facilities located nearby should be taken into account in assessing the overall impact of air Hg on U-Hg in this area.

Spatial mapping of ozone and SO2 trends in Europe.

This article investigates the spatial mapping of temporal trends in air quality for all of Europe. Such spatially distributed maps provide information for policy making and for understanding the spatial character of air quality trends. Previous trend studies have concentrated on individual, or groups of, monitoring sites looking at the trends of these. In this study use is made of statistical interpolation methods that combine observed and modelled data in an optimised way. Log-normal residual kriging with multiple linear regression is used to produce annual maps of air quality indicators for ozone (AOT40; Accumulated Dose of Ozone Over a Threshold of 40 ppb) and SO(2) (annual mean) for the period 1996-2005. Trends in these maps are then calculated and their significance and uncertainty are assessed. The methodology is effectively used for mapping SO(2) trends to a significant level in most of Europe. However, trends in AOT40 are less clearly defined since the uncertainty is generally of the same order as, or greater than, the calculated trends. A general north to south gradient in AOT40 trends can be seen, with downward trends in the UK and Scandinavia but upward trends in the Mediterranean region.