Deep Learning-Based Estimation of Reverberant Environment for Audio Data Augmentation

This paper proposes an audio data augmentation method based on deep learning in order to improve the performance of dereverberation. Conventionally, audio data are augmented using a room impulse response, which is artificially generated by some methods, such as the image method. The proposed method estimates a reverberation environment model based on a deep neural network that is trained by using clean and recorded audio data as inputs and outputs, respectively. Then, a large amount of a real augmented database is constructed by using the trained reverberation model, and the dereverberation model is trained with the augmented database. The performance of the augmentation model was verified by a log spectral distance and mean square error between the real augmented data and the recorded data. In addition, according to dereverberation experiments, the proposed method showed improved performance compared with the conventional method.

Statistical Hauser-Feshbach Model Description of (n,α) Reaction Cross Sections for the Weak s-Process

The (n,α) reaction contributes in many processes of energy generation and nucleosynthesis in stellar environment. Since experimental data are available for a limited number of nuclei and in restricted energy ranges, at present only theoretical studies can provide predictions for all astrophysically relevant (n,α) reaction cross sections. The purpose of this work is to study (n,α) reaction cross sections for a set of nuclei contributing in the weak s-process nucleosynthesis. Theory framework is based on the statistical Hauser-Feshbach model implemented in TALYS code with nuclear masses and level densities based on Skyrme energy density functional. In addition to the analysis of the properties of calculated (n,α) cross sections, the Maxwellian averaged cross sections are described and analyzed for the range of temperatures in stellar environment. Model calculations determined astrophysically relevant energy windows in which (n,α) reactions occur in stars. In order to reduce the uncertainties in modeling (n,α) reaction cross sections for the s-process, novel experimental studies are called for. Presented results on the effective energy windows for (n,α) reaction in weak s-process provide a guidance for the priority energy ranges in the future experimental studies.

A holistic framework to examine the impact of user, organizational and data factors on the use of big data analytics systems

Introduction. More than 80% of big data projects have failed to meet project sponsors' expectations. This study adopts the technology-organization-environment model to provide a holistic framework to examine the key factors contributing to the success of business analytics projects. Method. The paper develops a survey questionnaire consisting of thirty items on eight constructs, based on previous studies from cognitive-experiential self-efficacy, information technology mindfulness, technology-organization-environment, and information systems success theories. A total of 236 actual users in twenty organizations participated in the study. Analysis. We employed structural equation modelling with partial least squares to test the proposed hypotheses. Results. Our analysis shows that openness to novelty, the rational cognitive thinking style, organizational compatibility, organizational readiness, and data connectivity have a positive effect on the active use of big data analytics systems. Data compatibility does not emerge as a strong antecedent for frequency or duration. The active use of these types of applications is a mediating factor that can significantly improve individual job performance. Our findings provide insights into the user’s relative value, as well as organizational and data factors that can be used to promote big data analytics systems in an organization. Conclusions. Our findings strongly suggest that such an integrative approach can help an organization understand the challenges of promoting big data analytics to use and improve employees' job performance.

CONTROL OF REMOTELY OPERATED UNDERWATER VEHICLE USING MODEL PREDICTIVE CONTROL

This paper focuses on application of model predictive control on attitude control of remotely operated underwater vehicle. These vehicles are used in scientific, defence and oceanography applications. Remotely Operated Vehicle (ROV) considered in this paper is nonlinear model and complex. MPC is applied on ROV model to track in desired set point trajectories in the presence of uncertainties. Simulation has been carried out in MATLAB environment. Model Predictive Control has given significantly good results compared to PID, Adaptive and Variable structure control.

Thermal modelling and simulation of a laboratory building for cooling energy consumption

The main objective of this study is to identify and calculate the necessary variables for a building that can be used as input data in energy simulation tools and to determine the cooling energy consumption buildings and to examine the influence of applying energy rationalization measures. Based on that, we developed, applied, and validated a Simulink-based energy demand model of a governmental building to quantify the amount of cooling energy consumption in a case study environment. The model includes an air conditioner (plant model), controlled by a thermostat (controller model), to cool the laboratory building (environment model) to the desired temperature. The processes used here to construct the model layout and algorithm design can be extended to accommodate other multipart models or different construction sectors.

A Manipulator Control Method Based on Deep Deterministic Policy Gradient with Parameter Noise

Focusing on the motion control problem of two link manipulator, a manipulator control approach based on deep deterministic policy gradient with parameter noise is proposed. Firstly, the manipulator simulation environment is built. And then the three deep reinforcement learning models named the deep deterministic policy gradient (DDPG), asynchronous advantage actor-critical (A3C) and distributed proximal policy optimization (DPPO) are established for training according to the target setting, state variables and reward & punishment mechanism of the environment model. Finally the motion control of two link manipulator is realized. After comparing and analyzing the three models, the DDPG approach based on parameter noise is proposed for further research to improve its applicability, so as to cut down the debugging time of the manipulator model and reach the goal smoothly. The experimental results indicate that the DDPG approach based on parameter noise can control the motion of two link manipulator effectively. The convergence speed of the control model is significantly promoted and the stability after convergence is improved. In comparison with the traditional control approach, the DDPG control approach based on parameter noise has higher efficiency and stronger applicability.

Formation of a value attitude to health among younger school children in an innovative educational environment

The purpose of the research: to study the substantive foundations of the organization of health education and health savings in younger schoolchildren; to develop a "Health saving space of a child in an innovative educational environment" model; to evaluate the effectiveness of the use of an educational project. The study involved 44 children from the 2nd grade (Gymnasium No. 1 of the city of Krasnozmensk, Moscow region); 44 parents; scientific and pedagogical staff. Based on the study of scientific literature, theoretical analysis and systematization of domestic experience on the problem of value attitude to the health of younger schoolchildren have been carried out. The following methods were used in the course of the study: the observation method; the author's questionnaire (the study of the attitude of younger schoolchildren to their health); "The methodology of the study of value orientations" (M. Rokich). The conducted empirical study confirmed the hypothesis about the need to form a value-based attitude to health in children. The research is a necessary direction in the development of preventive pedagogy and psychology on health education and health saving of children. The materials of the article can be used by teachers, psychologists, teachers, students.

Mastering continuous improvement (CI): the roles and competences of mid-level management and their impact on the organisation's CI capability

PurposeThis paper establishes a comprehensive basis for understanding the roles and competences of mid-level management and their influence on the effectiveness of continuous improvement (CI) capability within an organisation.Design/methodology/approachThis research builds upon the hypothesis that methods alone do not lead to successful CI capability development. It focuses on the role of mid-level management in driving a CI environment that underpins the effectiveness of CI capability. A reference model for the CI environment is synthesised based on critical literature review, integrating CI culture, CI enablers and CI leadership elements. A comprehensive framework is introduced to define CI leadership roles and competence indicators. A quantitative benchmarking study involving structured interviews with 15 UK organisations was undertaken to collect evidence for a causal relationship between CI leadership competences and CI capability.FindingsAnalysis of the benchmarking data provides clear evidence of the causal relationship between the CI leadership competences of mid-level management and CI capability of the organisation. Given that the empirical study was structured on the basis of the CI leadership roles and competences framework introduced in this paper, this also provides validation for the proposed framework and the CI environment model.Practical implicationsThe evidence-based knowledge of the positive relationship between the mid-management CI leadership competences and the effectiveness of the CI capability informs strategic organisational development interventions towards enhancing CI capability and effectiveness, ultimately underpinning productivity enhancement and sustainability. The framework for mid-level management CI leadership roles, responsibilities and competences introduced in this paper and grounded in underpinning work undertaken within a large automotive Original Equipment Manufacturer (OEM), can be adapted by any organisation. The CI environment reference model should provide a comprehensive support for strategists to communicate the framework for CI capability improvement within an organisation, to enhance acceptability and adherence to improvement actions.Originality/valueThis research proves for the first time the significance of the causal relationship between the CI leadership competences and the effectiveness of the CI capability within an organisation, thus filling an important gap between established previous work, focussing on the role of mid-level management on one side and practitioner and team level roles, methodologies and tools. The proposed CI environment model is a theoretical contribution with reference value for both practice and further studies. The comprehensive framework for mid-level management CI leadership roles, responsibilities and competences introduced in this paper provides sound foundation to deliver CI leadership in the workplace.

Investigation of the implementation of a Communication Enhanced Environment model on an acute/slow stream rehabilitation and a rehabilitation ward: A before-and-after pilot study

Objectives: Develop and implement a Communication Enhanced Environment model and explore its effect on language activities for patients early after stroke. Method and design: Before-and-after pilot study. Setting: An acute/slow stream rehabilitation and a rehabilitation ward in a private hospital in Perth, Western Australia. Participants: Fourteen patients recruited within 21 days of stroke. Seven recruited during the before-phase (control group: patients with aphasia = 3, patients without aphasia = 4) and seven recruited in the after-phase (intervention group: patients with aphasia = 4, patients without aphasia = 3). Intervention: The intervention group exposed to a Communication Enhanced Environment model had access to equipment, resources, planned social activities and trained communication partners. Both groups received usual stroke care. Data collection: Hospital site champions monitored the availability of the intervention. Behavioural mapping completed during the first minute of each 5-minute interval over 12 hours (between 7 am and 7 pm) determined patient engagement in language activities. Results: Seventy-one percent of the Communication Enhanced Environment model was available to the intervention group who engaged in higher, but not significant (95% CI), levels of language activities (600 of 816 observation time points, 73%) than the control group (551 of 835 observation time points, 66%). Unforeseen reorganisation of the acute ward occurred during the study. Conclusions: Implementation of a Communication Enhanced Environment model was feasible in this specific setting and may potentially influence patients’ engagement in language activities. The unforeseen contextual challenges that occurred during the study period demonstrate the challenging nature of the hospital environment and will be useful in future research planning.