Search

In recent years, automatic speech recognition (ASR) models greatly improved transcription performance both in clean, low noise, acoustic conditions and in reverberant environments. However, all these systems rely on the availability of hundreds of hours of labelled training data in specific acoustic conditions. When such a training dataset is not available, the performance of the system is heavily impacted. For example, this happens when a specific acoustic environment or a particular population of speakers is under-represented in the training dataset. Specifically, in this paper we investigate the effect of accented speech data on an off-the-shelf ASR system. Furthermore, we suggest a strategy based on zero-shot text-to-speech to augment the accented speech corpora. We show that this augmentation method is able to mitigate the loss in performance of the ASR system on accented data up to 5% word error rate reduction (WERR). In conclusion, we demonstrate that by incorporating a modest fraction of real with synthetically generated data, the ASR system exhibits superior performance compared to a model trained exclusively on authentic accented speech with up to 14% WERR., Comment: Accepted to the Asilomar 2023 Conference

The growing interest in cybersecurity has significantly increased articles designing and implementing various Cyber Deception (CYDEC) mechanisms. This trend reflects the urgent need for new strategies to address cyber threats effectively. Since its emergence, CYDEC has established itself as an innovative defense against attackers, thanks to its proactive and reactive capabilities, finding applications in numerous real-life scenarios. Despite the considerable work devoted to CYDEC, the literature still presents significant gaps. In particular, there has not been (i) a comprehensive analysis of the main components characterizing CYDEC, (ii) a generic classification covering all types of solutions, nor (iii) a survey of the current state of the literature in various contexts. This article aims to fill these gaps through a detailed review of the main features that comprise CYDEC, developing a comprehensive classification taxonomy. In addition, the different frameworks used to generate CYDEC are reviewed, presenting a more comprehensive one. Existing solutions in the literature using CYDEC, both without Artificial Intelligence (AI) and with AI, are studied and compared. Finally, the most salient trends of the current state of the art are discussed, offering a list of pending challenges for future research., Comment: 38 pages

Recordings in everyday life require privacy preservation of the speech content and speaker identity. This contribution explores the influence of noise and reverberation on the trade-off between privacy and utility for low-cost privacy-preserving methods feasible for edge computing. These methods compromise spectral and temporal smoothing, speaker anonymization using the McAdams coefficient, sampling with a very low sampling rate, and combinations. Privacy is assessed by automatic speech and speaker recognition, while our utility considers voice activity detection and speaker diarization. Overall, our evaluation shows that additional noise degrades the performance of all models more than reverberation. This degradation corresponds to enhanced speech privacy, while utility is less deteriorated for some methods., Comment: Accepted for publication at IWAENC 2024

We propose to build a Flexible Stellarator Physics Facility to explore promising regions of the vast parameter space of disruption-free stellarator solutions for Fusion Pilot Plants (FPPs)., Comment: White paper submitted to FESAC subcommittee on Facilities, 8 pages

An integrated modeling framework for investigating the application of solid boron powder injection for real-time surface conditioning of plasma-facing components in tokamak environments is presented. Utilizing the DIII-D impurity powder dropper setup, this study simulates B powder injection scenarios ranging from mg/s to tens of mg/s, corresponding to B flux rates of $10^{20}-10^{21}$ B/s in standard L-mode conditions. The comprehensive modeling approach combines EMC3-EIRENE for simulating the D plasma background and DIS for the ablation and transport of the B powder particles. The results show substantial transport of B to the inboard lower divertor, predominantly influenced by the main ion plasma flow. The dependency on powder particle size (5-250 $\mu$m) was found to be insignificant for the scenario considered. The effects of erosion and redeposition were considered to reconcile the discrepancies with experimental observations, which saw substantial deposition on the outer divertor PFCs. For this purpose, the WallDYN3D code was updated to include B sources within the plasma domain and integrated into the modeling framework. The mixed-material migration modeling shows evolving B deposition patterns, suggesting the formation of mixed B-C layers or predominantly B coverage depending on the powder mass flow rate. While the modeling outcomes at lower B injection rates tend to align with experimental observations, the prediction of near-pure B layers at higher rates has yet to be experimentally verified in the C environment of the DIII-D tokamak. The extensive reach of B layers found in the modeling suggests the need for modeling that encompasses the entire wall geometry for more accurate experimental correlations. This integrated approach sets a precedent for analyzing and applying real-time in-situ boron coating techniques in advanced tokamak scenarios, potentially extendable to ITER.

One way to shed some light on the star formation process this process is to analyse the relationship between the spatial distributions of gas and newly formed stars. In order to obtain robust results, it is necessary for this comparison to be made using quantitative and consistent descriptors applied to the same star-forming region. Here, we use fractal analysis to characterise and compare in a self-consistent way the structure of the cloud and the distribution of young stellar objects (YSO) in the Dragonfish star-forming complex. We used different emission maps of the Dragonfish Nebula and photometric information from the AllWISE catalogue to select a total of 1082 YSOs in the region, for some of which we derived physical properties from their spectral energy distributions (SEDs). For both datasets (cloud images and YSOs), the three-dimensional fractal dimension (Df) was calculated using previously developed and calibrated algorithms. The fractal dimension of the Dragonfish Nebula (Df = 2.6-2.7) agrees very well with values previously obtained for the Orion, Ophiuchus, and Perseus clouds. On the other hand, YSOs exhibit on average a significantly smaller value (Df = 1.9-2.0) that indicates a much more clumpy structure than the material from which they formed. This is a clear and direct evidence that the clustering degree of the newly born stars is significantly higher than that of the parent cloud from which they formed, but the physical mechanism behind this behaviour is still not clear. Additionally, younger Class I and Class II sources have smaller values (Df = 1.7 +/- 0.1) than more evolved Transition Disk objects (Df = 2.2 +/- 0.1), evidencing a certain evolutionary effect where an initially clumpy structure tends to gradually disappear over time., Comment: 10 pages, 9 figures, 4 tables, submitted to A&A

Cybersecurity is developing rapidly, and new methods of defence against attackers are appearing, such as Cyber Deception (CYDEC). CYDEC consists of deceiving the enemy who performs actions without realising that he/she is being deceived. This article proposes designing, implementing, and evaluating a deception mechanism based on the stealthy redirection of TCP communications to an on-demand honey server with the same characteristics as the victim asset, i.e., it is a clone. Such a mechanism ensures that the defender fools the attacker, thanks to stealth redirection. In this situation, the attacker will focus on attacking the honey server while enabling the recollection of relevant information to generate threat intelligence. The experiments in different scenarios show how the proposed solution can effectively redirect an attacker to a copied asset on demand, thus protecting the real asset. Finally, the results obtained by evaluating the latency times ensure that the redirection is undetectable by humans and very difficult to detect by a machine.

It is undeniable that we are witnessing an unprecedented digital revolution. However, recent years have been characterized by the explosion of cyberattacks, making cybercrime one of the most profitable businesses on the planet. That is why training in cybersecurity is increasingly essential to protect the assets of cyberspace. One of the most vital tools to train cybersecurity competencies is the Cyber Range, a virtualized environment that simulates realistic networks. The paper at hand introduces SCORPION, a fully functional and virtualized Cyber Range, which manages the authoring and automated deployment of scenarios. In addition, SCORPION includes several elements to improve student motivation, such as a gamification system with medals, points, or rankings, among other elements. Such a gamification system includes an adaptive learning module that is able to adapt the cyberexercise based on the users' performance. Moreover, SCORPION leverages learning analytics that collects and processes telemetric and biometric user data, including heart rate through a smartwatch, which is available through a dashboard for instructors. Finally, we developed a case study where SCORPION obtained 82.10% in usability and 4.57 out of 5 in usefulness from the viewpoint of a student and an instructor. The positive evaluation results are promising, indicating that SCORPION can become an effective, motivating, and advanced cybersecurity training tool to help fill current gaps in this context., Comment: 31 pages

There is no denying that the use of Information Technology (IT) is undergoing exponential growth in today's world. This digital transformation has also given rise to a multitude of security challenges, notably in the realm of cybercrime. In response to these growing threats, public and private sectors have prioritized the strengthening of IT security measures. In light of the growing security concern, Artificial Intelligence (AI) has gained prominence within the cybersecurity landscape. This paper presents a comprehensive survey of recent advancements in AI-driven threat response systems. To the best of our knowledge, the most recent survey covering the AI reaction domain was conducted in 2017. Since then, considerable literature has been published, and therefore, it is worth reviewing it. In this comprehensive survey of the state of the art reaction systems, five key features with multiple values have been identified, facilitating a homogeneous comparison between the different works. In addition, through a meticulous methodology of article collection, the 22 most relevant publications in the field have been selected. Then each of these publications has been subjected to a detailed analysis using the features identified, which has allowed for the generation of a comprehensive overview revealing significant relationships between the papers. These relationships are further elaborated in the paper, along with the identification of potential gaps in the literature, which may guide future contributions. A total of seven research challenges have been identified, pointing out these potential gaps and suggesting possible areas of development through concrete proposals.

The proliferation of car sharing services in recent years presents a promising avenue for advancing sustainable transportation. Beyond merely reducing car ownership rates, these systems can play a pivotal role in bolstering grid stability through the provision of ancillary services via vehicle-to-grid (V2G) technologies - a facet that has received limited attention in previous research. In this study, we analyze the potential of V2G in car sharing by designing future scenarios for a national-scale service in Switzerland. We propose an agent-based simulation pipeline that considers population changes as well as different business strategies of the car sharing service, and we demonstrate its successful application for simulating scenarios for 2030. To imitate car sharing user behavior, we develop a data-driven mode choice model. Our analysis reveals important differences in the examined scenarios, such as higher vehicle utilization rates for a reduced fleet size as well as in a scenario featuring new car sharing stations. These disparities translate into variations in the power flexibility of the fleet available for ancillary services, ranging from 12 to 50 MW, depending on the scenario and the time of the day. Furthermore, we conduct a case study involving a subset of the car sharing fleet, incorporating real-world electricity pricing data. The case study substantiates the existence of a sweet spot involving monetary gains for both power grid operators and fleet owners. Our findings provide guidelines to decision makers and underscore the pressing need for regulatory enhancements concerning power trading within the realm of car sharing.

This article presents the affordances that Generative Artificial Intelligence can have in misinformation and disinformation contexts, major threats to our digitalized society. We present a research framework to generate customized agent-based social networks for disinformation simulations that would enable understanding and evaluating the phenomena whilst discussing open challenges.

The vast majority of approaches to speaker anonymization involve the extraction of fundamental frequency estimates, linguistic features and a speaker embedding which is perturbed to obfuscate the speaker identity before an anonymized speech waveform is resynthesized using a vocoder. Recent work has shown that x-vector transformations are difficult to control consistently: other sources of speaker information contained within fundamental frequency and linguistic features are re-entangled upon vocoding, meaning that anonymized speech signals still contain speaker information. We propose an approach based upon neural audio codecs (NACs), which are known to generate high-quality synthetic speech when combined with language models. NACs use quantized codes, which are known to effectively bottleneck speaker-related information: we demonstrate the potential of speaker anonymization systems based on NAC language modeling by applying the evaluation framework of the Voice Privacy Challenge 2022., Comment: Accepted at ICASSP 2024

The analysis of conversations recorded in everyday life requires privacy protection. In this contribution, we explore a privacy-preserving feature extraction method based on input feature dimension reduction, spectral smoothing and the low-cost speaker anonymization technique based on McAdams coefficient. We assess the utility of the feature extraction methods with a voice activity detection and a speaker diarization system, while privacy protection is determined with a speech recognition and a speaker verification model. We show that the combination of McAdams coefficient and spectral smoothing maintains the utility while improving privacy., Comment: Submitted to ITG Conference on Speech Communication, 2023

In recent years, the need for privacy preservation when manipulating or storing personal data, including speech , has become a major issue. In this paper, we present a system addressing the speaker-level anonymization problem. We propose and evaluate a two-stage anonymization pipeline exploiting a state-of-the-art anonymization model described in the Voice Privacy Challenge 2022 in combination with a zero-shot voice conversion architecture able to capture speaker characteristics from a few seconds of speech. We show this architecture can lead to strong privacy preservation while preserving pitch information. Finally, we propose a new compressed metric to evaluate anonymization systems in privacy scenarios with different constraints on privacy and utility., Comment: submitted to INTERSPEECH

This paper proposes a general and practical approach to estimate the economic benefits of optimally controlling deferrable loads in a Distribution System Operator's (DSO) grid, without relying on historical observations. We achieve this by learning the simulated response of flexible loads to random control signals, using a non-parametric global forecasting model. An optimal control policy is found by including the latter in an optimization problem. We apply this method to electric water heaters and heat pumps operated through ripple control and show how flexibility, including rebound effects, can be characterized and controlled. Finally, we show that the forecaster's accuracy is sufficient to completely bypass the simulations and directly use the forecaster to estimate the economic benefit of flexibility control.

Any audio recording encapsulates the unique fingerprint of the associated acoustic environment, namely the background noise and reverberation. Considering the scenario of a room equipped with a fixed smart speaker device with one or more microphones and a wearable smart device (watch, glasses or smartphone), we employed the improved proportionate normalized least mean square adaptive filter to estimate the relative room impulse response mapping the audio recordings of the two devices. We performed inter-device distance estimation by exploiting a new set of features obtained extending the definition of some acoustic attributes of the room impulse response to its relative version. In combination with the sparseness measure of the estimated relative room impulse response, the relative features allow precise inter-device distance estimation which can be exploited for tasks such as best microphone selection or acoustic scene analysis. Experimental results from simulated rooms of different dimensions and reverberation times demonstrate the effectiveness of this computationally lightweight approach for smart home acoustic ranging applications

Deploying real-time control on large-scale fleets of electric vehicles (EVs) is becoming pivotal as the share of EVs over internal combustion engine vehicles increases. In this paper, we present a Vehicle-to-Grid (V2G) algorithm to simultaneously schedule thousands of EVs charging and discharging operations, that can be used to provide ancillary services. To achieve scalability, the monolithic problem is decomposed using the alternating direction method of multipliers (ADMM). Furthermore, we propose a method to handle bilinear constraints of the original problem inside the ADMM iterations, which changes the problem class from Mixed-Integer Quadratic Program (MIQP) to Quadratic Program (QP), allowing for a substantial computational speed up. We test the algorithm using real data from the largest carsharing company in Switzerland and show how our formulation can be used to retrieve flexibility boundaries for the EV fleet. Our work thus enables fleet operators to make informed bids on ancillary services provision, thereby facilitating the integration of electric vehicles.

Can a robot be a personal dating coach? Even with the increasing amount of conversational data on the internet, the implementation of conversational robots remains a challenge. In particular, a detailed and professional counseling log is expensive and not publicly accessible. In this paper, we develop a robot dating coaching system leveraging corpus from online communities. We examine people's perceptions of the dating coaching robot with a dialogue module. 97 participants joined to have a conversation with the robot, and 30 of them evaluated the robot. The results indicate that participants thought the robot could become a dating coach while considering the robot is entertaining rather than helpful.

The ability to forecast the concentration of air pollutants in an urban region is crucial for decision-makers wishing to reduce the impact of pollution on public health through active measures (e.g. temporary traffic closures). In this study, we present a machine learning approach applied to the forecast of the day-ahead maximum value of the ozone concentration for several geographical locations in southern Switzerland. Due to the low density of measurement stations and to the complex orography of the use case terrain, we adopted feature selection methods instead of explicitly restricting relevant features to a neighbourhood of the prediction sites, as common in spatio-temporal forecasting methods. We then used Shapley values to assess the explainability of the learned models in terms of feature importance and feature interactions in relation to ozone predictions; our analysis suggests that the trained models effectively learned explanatory cross-dependencies among atmospheric variables. Finally, we show how weighting observations helps in increasing the accuracy of the forecasts for specific ranges of ozone's daily peak values.

While social media has been proved as an exceptionally useful tool to interact with other people and massively and quickly spread helpful information, its great potential has been ill-intentionally leveraged as well to distort political elections and manipulate constituents. In the paper at hand, we analyzed the presence and behavior of social bots on Twitter in the context of the November 2019 Spanish general election. Throughout our study, we classified involved users as social bots or humans, and examined their interactions from a quantitative (i.e., amount of traffic generated and existing relations) and qualitative (i.e., user's political affinity and sentiment towards the most important parties) perspectives. Results demonstrated that a non-negligible amount of those bots actively participated in the election, supporting each of the five principal political parties.

Gradient boosted trees are competition-winning, general-purpose, non-parametric regressors, which exploit sequential model fitting and gradient descent to minimize a specific loss function. The most popular implementations are tailored to univariate regression and classification tasks, precluding the possibility of capturing multivariate target cross-correlations and applying structured penalties to the predictions. In this paper, we present a computationally efficient algorithm for fitting multivariate boosted trees. We show that multivariate trees can outperform their univariate counterpart when the predictions are correlated. Furthermore, the algorithm allows to arbitrarily regularize the predictions, so that properties like smoothness, consistency and functional relations can be enforced. We present applications and numerical results related to forecasting and control.