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.

Effect of Tempo on Temporal Expectation Driven by Rhythms in Dual-Task Performance

Temporal expectation is the ability to focus attention at a particular moment in time to optimize performance, which has been shown to be driven by regular rhythms. However, whether the rhythm-based temporal expectations rely upon automatic processing or require the involvement of controlled processing has not been clearly established. Furthermore, whether the mechanism is affected by tempo remains unknown. To investigate this research question, the present study used a dual-task procedure. In a single task, the participants were instructed to respond to a visual target preceded by a regular or an irregular visual rhythm under a fast (500 ms) or slow (3,500 ms) tempo. The dual-task simultaneously combined a working memory (WM) task. The results showed temporal expectation effects in which the participants responded faster to the regular than to the irregular conditions in a single task. Moreover, this effect persisted under dual-task interference in the fast tempo condition but was impaired in the slow tempo condition. These results revealed that rhythmic temporal expectation induced by fast tempo was dependent on automatic processing. However, compared with the faster tempo, temporal expectation driven by a slower tempo might involve more controlled processing.

An all-sky camera images classification method using cloud cover features

The all-sky camera (ASC) images can reflect the local cloud cover information, and the cloud cover is one of the first factors considered for astronomical observatory site selection. Therefore, the realization of automatic classification of the ASC images plays an important role in astronomical observatory site selection. In this paper, three cloud cover features are proposed for the TMT (Thirty Meter Telescope) classification criteria, namely cloud weight, cloud area ratio and cloud dispersion. After the features are quantified, four classifiers are used to recognize the classes of the images. Four classes of ASC images are identified: “Clear”, “Inner”, “Outer” and “Covered”. The proposed method is evaluated on a large dataset, which contains 7328 ASC images taken by an all-sky camera located in Xinjiang (38.19° N, 74.53° E). In the end, the method achieves an accuracy of 97.28 % and F1_score of 96.97 % by a random forest (RF) classifier, which greatly improves the efficiency of automatic processing of the ASC images.

Automatic Processing of the Information Received from the Space Rocket Onboard Video Control System.

The paper discusses the information processing method developed by the Mis-sion Control Center (AO TSNIImash) specialists, used to process the infor-mation received from the Soyuz space rocket onboard video control system. Specific features of the information being received are presented. Methods of restoring corrupt information are proposed. The results of information pro-cessing during real Soyuz-FG and Soyuz-2 space rocket launches are given.

EXPRESS: Sometimes nothing is simply nothing: Automatic processing of empty sets

Previous work using the numerical comparison task has shown that an empty set, the nonsymbolic manifestation of zero, can be represented as the smallest quantity of the numerical magnitude system. In the present study, we examined whether an empty set can be represented as such under conditions of automatic processing in which deliberate processing of stimuli magnitudes is not required by the task. In Experiment 1, participants performed physical and numerical comparisons of empty sets (i.e., empty frames) and of other numerosities presented as framed arrays of 1 to 9 dots. The physical sizes of the frames varied within pairs. Both tasks revealed a size congruity effect (SCE) for comparisons of non-empty sets. In contrast, comparisons to empty sets produced an inverted SCE in the physical comparison task, while no SCE was found for comparisons to empty sets in the numerical comparison task. In Experiment 2, participants performed an area comparison task using the same stimuli as Experiment 1 to examine the effect of visual cues on the automatic processing of empty sets. The results replicated the findings of the physical comparison task in Experiment 1. Taken together, our findings indicate that empty sets are not perceived as “zero”, but rather as “nothing”, when processed automatically. Hence, the perceptual dominance of empty sets seems to play a more important role under conditions of automatic processing, making it harder to abstract the numerical meaning of zero from empty sets.