Automatic Processing Pipeline for Collecting and Annotating Air-Traffic Voice Communication Data

This document describes our pipeline for automatic processing of ATCO pilot audio communication we developed as part of the ATCO2 project. So far, we collected two thousand hours of audio recordings that we either preprocessed for the transcribers or used for semi-supervised training. Both methods of using the collected data can further improve our pipeline by retraining our models. The proposed automatic processing pipeline is a cascade of many standalone components: (a) segmentation, (b) volume control, (c) signal-to-noise ratio filtering, (d) diarization, (e) ‘speech-to-text’ (ASR) module, (f) English language detection, (g) call-sign code recognition, (h) ATCO—pilot classification and (i) highlighting commands and values. The key component of the pipeline is a speech-to-text transcription system that has to be trained with real-world ATC data; otherwise, the performance is poor. In order to further improve speech-to-text performance, we apply both semi-supervised training with our recordings and the contextual adaptation that uses a list of plausible callsigns from surveillance data as auxiliary information. Downstream NLP/NLU tasks are important from an application point of view. These application tasks need accurate models operating on top of the real speech-to-text output; thus, there is a need for more data too. Creating ATC data is the main aspiration of the ATCO2 project. At the end of the project, the data will be packaged and distributed by ELDA.

ARTIFICIAL LEARNING-BASED PROCTORING SOLUTION FOR REMOTE ONLINE ASSESSMENTS: “VPROCTOR”

Technological developments have paved the way for courses, trainings and assessments to be made online remotely in education and employment. In the meantime, over the years, the demand for online distance learning has increased rapidly. Eventually, it has been seen that this is a necessity and it has become widespread during the COVID-19 pandemic period. As can be seen in the example of massive online open courses, although remote learning is carried out online, perhaps the most important problem is how to evaluate the relevant courses safely and reliably. Thus, remote online proctoring is becoming an increasingly popular and needed issue. In this study, the issue of conducting the examination remotely online via internet-connected video and audio communication will be evaluated. Furthermore, the solution called vProctor was developed to contribute to the elimination of deficiencies in remote online proctoring using artificial learning. Overall, it has been observed that the proposed solution is managed to determine inappropriate behaviors like cheating in the online assessments.

Sistem Komunikasi Audio dengan Teknologi Visible Light Communication (VLC) Menggunakan Laser Led

Kegiatan masyarakat di era sekarang yang serba digital menggunakan teknologi tanpa kabel (wireless). Teknologi tanpa kabel menjadi andalan komunikasi, namun seiring perkembangann zaman, ada sebuah teknologi tanpa kabel dengan memanfaatkan cahaya tampak. Teknologi ini dikenal dengan istilah visible light communication (VLC). Sistem VLC terdiri dari bagian transmitter dan receiver. Transmitter berupa komponen-komponen elektronika yang terintegrasi dengan cahaya tampak sebagai media transmisi. Pada penelitian ini membuat bagian transmitter dan receiver untuk komunikasi audio. Cahaya tampak sebagai media transmisi menggunakan laser led karena cahaya dari laser led ini lebih fokus dan pancaran cahayanya lebih jauh dibandingkan dengan yang lain. Sedangkan di sisi receiver menggunakan solar panel sebagai penerima sinyal informasi yang dikirimkan. Penelitian ini berhasil membuat sebuah sistem komunikasi audio berbasis visible light communication. Variasi jarak antara transmitter dan receiver mempengaruhi kualitas suara yang diterima. Pada jarak 500 cm kualitas suara yaitu 99,2 dB pada kondisi gelap dan 99,1 dB kondisi terang. Perubahan sudut pancar transmitter terhadap receiver juga mempengaruhi hasil kualitas suara. Kualitas suara terbaik pada sudut 900 di kondisi terang dengan jarak 50 cm sebesar 97,9 dB sedangkan di kondisi gelap dengan jarak 50 cm pada sudut 1000 sebesar 98,1 dB. Abstract The people activities in today's digital era use wireless technology. Wireless technology has become the mainstay of communication, but along with the times, there is a wireless technology that uses visible light. This technology is known as visible light communication (VLC). The VLC system consist of transmitter and receiver. Transmitters are electronic components that are integrated with visible light as a transmission medium. In this study, the transmitter and receiver are made for audio communication. Visible light as a transmission medium uses an LED laser because the light from this LED laser is more focused and the light beam is farther away than the others. While on the receiver side using a solar panel as a receiver of the information signal that is sent. This research succeeded in making an audio communication system based on visible light communication. Variations in the distance between the transmitter and receiver affect the quality of the received sound. At a distance of 500 cm in dark conditions, the sound quality is 99.2 dB in light conditions and 99.1 dB. Changes in the beam angle of tranmitter to receiver also affect the sound quality. The best sound quality at an angle of 900 in bright conditions with a distance of 50 cm is 97.9 dB while in dark conditions with a distance of 50 cm at an angle of 1000 is 98.1 dB.

Management of patients with atherosclerotic lower extremity peripheral arterial disease in the outpatient setting

Aim. To analyze and scientifically substantiate the need for the creation of a program for the outpatient observation with COZANC using remote patronage.Methods. The study prospectively included 150 patients with atherosclerosis of the arteries of the lower extremities. The average age of the patients was 68.3 ± 9.7 years. Clinical and anamnestic data were studied, and patients were interviewed using questionnaires and scales. Two observation groups were formed: the main one consisted of 70 patients, who were included in the program of remote patronage clinical examination, and 80 patients were under the standard model of outpatient observation with COZANC.Results. For a 3-month period of outpatient observation, an audio communication session was carried out for 70 patients, once a month. During the reporting period, no acute cardiovascular events were recorded in the main group, while in the comparison group there were 3 (3.75%) myocardial infarctions. The development of CLI occurred in the main group in 1 case, while in the comparison group there were 4 (5%) patients, in one case the lower limb was amputated. In the comparison group, there were 3 ischemic events that resulted in a lethal outcome.Conclusion. For the first time, a personalized remote patronage program was created and introduced into clinical practice at the level of primary health care, as a tool for optimizing the clinical examination program for patients with COZANC. The effectiveness of the remote patronage program for patients with atherosclerosis of the arteries of the lower extremities is described in relation to the prevention of acute cardiovascular events and the progression of COZANC.

Traversing Anthropocentric Horizons With Her: Trans-Corporeal Surrogacy, Enchantment, and Disenchantment in Human-Machine Assemblage

This article aims to shed light on the intricacies that overturn McLuhan's vision of technologies as extensions or prosthetics of human capabilities when applied to human-machine communication (HMC). Human and nonhuman entities co-evolve on an equal agential footing, immersed in mediatized assemblages. Building on the concepts of Deleuze and Guattari, Bennett, and others, it theorizes HMC as a cycle of sonic enchantment, culminating in trans-corporeal surrogacy, disrupted by disenchantment, and started again through re-enchantment. A new materialist framework helps explain the process of posthuman HMC. It provides a close-textual and visual analysis of Spike Jonze's film Her (2013), in which a human develops a romantic relationship with his AI assistant. The aspects of vulnerability, neediness, authenticity, trust, and intimacy surpass the lure of real-time personalized audio communication. The paper argues that artificial intelligence acquires autonomous agency through the processes of enchantment and mutual surrogacy that decenter humans in mediatized assemblages.

Integrating the Webinar as a Tool to Support Adult Training: A Case Study on the Training for Entrepreneurs of SMEs in Thailand

The COVID-19 pandemic has presented significant challenges for the education and training sector. We could have witnessed the rise of technology for webinars which are tools to deliver training and education through video and audio communication in the form of distance learning for instructors and participants. How-ever, commercial webinars such as zoom or webex may lack a management system for organizing the amount of participants for long term courses. The present study aimed to develop a webinar tool for the context of adult training for entrepreneurs of SMEs in Thailand. In order to develop a webinar tool, we proposed a general framework which consisted of three stages. The requirement stage aimed to explore the baseline survey from 411 participants and summarize the feature list of the webinar. The development stage employed the kanban methodology to develop each feature list and proposed the architecture of the system. Finally, the evaluation stage compares two groups of 110 participants between our webinar approach and zoom application with statistics of attendance regarding those who attended the course and also contains a satisfaction survey. The results show that the important feature to engage participants in long term courses were system notifications and the availability of a web-based platform for providing easy access to webinar.

A Feasibility Trial of HoloLens 2™; Using Mixed Reality Headsets to Deliver Remote Bedside Teaching During COVID-19

Background COVID-19 has had a catastrophic impact measured in human lives. Medical education has also been impacted: appropriately stringent infection control policies have precluded medical trainees from attending clinical teaching. Lecture-based education has been easily transferred to a digital platform, but bedside teaching has not transitioned so well. The aim of this study was to assess the feasibility of using a mixed reality (MR) headset to deliver remote bedside teaching to trainee doctors during the COVID-19 pandemic.Methods Two MR sessions were led by senior specialty registrars (the clinical trainers) wearing the HoloLens™ headset. The trainers selected patients admitted with pathology requiring their specialist input; thus, the educational sessions were opportunistic. The headset allowed bi-directional video and audio communication between the trainer and trainee doctors during the consultation. Trainee doctor conceptions of bedside teaching, impact of COVID-19 on bedside teaching and technical success of the session was evaluated using pre- and post-round questionnaires using 1 (‘strongly disagree’ or ‘never’) to 7 (‘strongly agree’ or ‘always’) Likert scales and white space questions. Feedback on acceptability of the round was collected verbally from patients after each encounter. Data related to clinician exposure to at risk patients and use of PPE were also collected.Results Pre-questionnaire respondents (n=24) strongly agreed that bedside teaching is key to educating clinicians (7, IQR 6-7). It was also apparent that bedside teaching had become a rarity during COVID-19 (2, IQR 2-4). Session 1 feedback (n=6) was adversely affected by a loose microphone connection. With the issue rectified for session 2 (n=4), most respondents strongly agreed that they felt like they were physically present for the session (7, IQR 6.75-7). Mixed-reality versus in-person teaching led to a 79.5% reduction in cumulative clinician exposure time and 83.3% reduction in PPE use. Conclusions This study is proof of principle that HoloLens™ can be used effectively to deliver high-quality clinical bedside teaching. This novel format confers significant advantages in terms of: minimising exposure of trainees to COVID-19; saving PPE; enabling larger attendance; and convenient accessible real-time clinical training.

Supervisor's capability and aptitude to supervise Health and Safety (H&S) induction training to site workforces: A case study of the construction industry of Pakistan

Health and Safety (H&S) is a professional approach to substitute harmless and vigorous work conditions for any industry. The construction industry is of utmost importance because it is associated with elevated accident rates compared to the other sectors. To some extent, cultural progression is also reliant on communication among all the groups inside a corporation. Effective communication happens in a construction corporation, among upper management and the employees arbitrated by the supervisors. Therefore, if the construction supervisor's capabilities are ignored, the plans to progress the satisfactory H&S culture may be unsuccessful. The aptitude for conducting H&S induction training to employees is one of the construction supervisor's abilities, and in developing countries, the supervisor's capabilities are ignored. Moreover, the organization's eagerness on H&S, which corresponds to the induction training, is relatively less. The current research reports an examination to analyze how the supervisors possess H&S capabilities and skills to oversee site H&S induction training to workforces successfully. For the analysis, Pakistan's construction industry was targeted, and numerous construction firms were contacted in this regard. The data for the intended objective was accumulated through talks and surveys with site supervisors in construction companies of various extents from all over Pakistan. The outcomes exposed insufficient training, an absence of management dedication, and an inadequate familiarity or usage of visual/audio communication assistance by construction companies in Pakistan. Thus, it is suggested that there is a need for the supervisors to understand the use of multilingual skills and current technologies to assist valuable site H&S induction training that can progress a path towards an optimistic H&S culture.