On‐device audio‐visual multi‐person wake word spotting

Audio‐visual wake word spotting is a challenging multi‐modal task that exploits visual information of lip motion patterns to supplement acoustic speech to improve overall detection performance.However,most audio‐visual wake word spotting models are only suitable for simple single‐speaker scenarios and require high computational complexity.Further development is hindered by complex multi‐person scenarios and computational limitations in mobile environments.In this paper,a novel audio‐visual model is proposed for on‐device multi‐person wake word spotting.Firstly,an attention‐based audio‐visual voice activity detection module is presented,which generates an attention score matrix of audio and visual representations to derive active speaker representation.Secondly,the knowledge distillation method is introduced to transfer knowledge from the large model to the on‐device model to control the size of our model.Moreover,a new audio‐visual dataset,PKU‐KWS,is collected for sentence‐level multi‐person wake word spotting.Experimental results on the PKU‐KWS dataset show that this approach outperforms the previous state‐of‐the‐art methods.

Visualization analysis of the capability of weapon system of systems for multi-dimensional indicators

In the field of weapon system of systems (WSOS) simulation, various indicators are widely used to describe the capability of WSOS, but it is always difficult to describe the comprehensive capability of WSOS quickly and intuitively by visualization of multi-dimensional indicators. A method of machine learning and visualization is proposed, which can display and analyze the capabilities of different WSOS in a two-dimensional plane. The analysis and comparison of the comprehensive capability of different components of WSOS is realized by the method, which consists of six parts: multiple simulations, key indicators mining, three spatial distance calculation, fusion project calculation, calculation of individual capability density, and calculation of multiple capability ranges overlay. Binding a simulation experiment, the collaborative analysis of six indicators and 100 possible kinds of red WSOS are achieved. The experimental results show that this method can effectively improve the quality and speed of capabilities analysis, reveal a large number of potential information, and provide a visual support for the qualitative and quantitative analysis model.

Visual analysis of the research hotspots of Neuman systems model in recent 20 years

Background:This paper uses information visualization software to sort out the relevant research of Neuman systems model at home and abroad in the past 20 years,and discusses the research hotspots and development trend in the field of Neuman systems model,so as to provide scientific and reliable reference for the future work and research.Methods:By using CiteSpace V software,this paper analyzes the literatures about the Neuman systems model collected in the core web of science database and CNKI database from 2001 to 2020,and analyzes the time distribution,research power distribution,research hotspots,research frontier and development trend of the Neuman systems model at home and abroad in the past 20 years.Results:The development trend of research in this field in foreign countries is relatively stable.The core strength of research is mainly in the United States,and the research hotspots are health,quality of life,caregivers,spirituality,etc;the research in this field in China is gradually on the rise,and there is no obvious research force,and the research hotspots are mainly quality of life,complications,anxiety,stressors,perioperative period,hypertension,etc.Conclusions:It has been proved that the model has a certain guiding effect on the development of nursing discipline in China.In China,there is still room for development in the research of this model.It is suggested that Chinese scholars can learn from foreign leading research forces to carry out in-depth research and expand its application scope.

Design and Simulation of an Audio Signal Alerting and Automatic Control System

A large part of our daily lives is spent with audio information. Massive obstacles are frequently presented by the colossal amounts of acoustic information and the incredibly quick processing times. This results in the need for applications and methodologies that are capable of automatically analyzing these contents. These technologies can be applied in automatic contentanalysis and emergency response systems. Breaks in manual communication usually occur in emergencies leading to accidents and equipment damage. The audio signal does a good job by sending a signal underground, which warrants action from an emergency management team at the surface. This paper, therefore, seeks to design and simulate an audio signal alerting and automatic control system using Unity Pro XL to substitute manual communication of emergencies and manual control of equipment. Sound data were trained using the neural network technique of machine learning. The metrics used are Fast Fourier transform magnitude, zero crossing rate, root mean square, and percentage error. Sounds were detected with an error of approximately 17%;thus, the system can detect sounds with an accuracy of 83%. With more data training, the system can detect sounds with minimal or no error. The paper, therefore, has critical policy implications about communication, safety, and health for underground mine.

Visualization of Approximate Reasoning Systems

ＶｉｓｕａｌｉｚａｔｉｏｎｏｆＡｐｐｒｏｘｉｍａｔｅＲｅａｓｏｎｉｎｇＳｙｓｔｅｍｓＺｈａｎｇＺｉｌｉ（ＤｅｐａｒｔｍｅｎｔｏｆＣｏｍｐｕｔｅｒＳｃｉｅｎｃｅ，ＳｏｕｔｈｗｅｓｔＣｈｉｎａＮｏｒｍａｌＵｎｉｖｅｒｓｉｔｙ，Ｃｈｏｎｇｑｉｎｇ６３０７１...

Visual System Simulator平台下锁相环综合分析与仿真

给出了基于VSS平台的锁相环仿真过程,并将其引入课堂教学和实验。同时给出了锁相环的仿真模型,构建了完整的锁相环系统,并给出了仿真波形与环路滤波器设计方法。实践证明,采用这种方式可以帮助学生对相关内容的理解,并为今后进行系统设计工作打下良好的基础。

用L-systems和遗传算法实现的叶子形状建模

阐述了组合两类技术的一种方法:用L-systems和遗传算法(GA)获取一种改写表达式描述叶子形状。一个L-systems用于构造给定改写表达式,而GA是用于改写表达式的适应参数。介绍了用目标函数实现实值参数替换方法,研究结果表明提出的方法产生可以接受的输 出效果。

Audio Signal Generator System Based On State Machines

A state machine can make program designing quicker,simpler and more efficient. This paper describes in detail the model for a state machine and the idea for its designing and gives the design process of the state machine through an example of audio signal generator system based on Labview. The result shows that the introduction of the state machine can make complex design processes more clear and the revision of programs easier.

APPLICATION OF MAP AND FILE INFORMATIONVISUALIZATION SYSTEM TO COMPREHENSIVEDIVISION OF NATURAL DISASTERS- Takingthe ChangjiangValleg as an Example

ABSTRACT: This paper generalizes the makeup and forming dynamic mechanism of natural disaster systems, principles and methods of comprehensive division of natural disasters, as well as structure, function and up-build routes of map and file information visualization system (MFIVS). Taking the Changjiang(Yangtze) Valley as an example, on the basis of revealing up the integrated mechanism on the formations of its natural disasters and its distributing law, thereafter, the paper relies on the MFIVS technique, adopts two top-down and bottom-up approaches to study a comprehensive division of natural disasters. It is relatively objective and precise that the required division results include three natural disaster sections and nine natural disaster sub-sections, which can not only provide a scientific basis for utilizing natural resources and controlling natural disaster and environmental degradation, but also be illuminated to a concise, practical and effective technique on comprehensive division.

Securing Technique Using Pattern-Based LSB Audio Steganography and Intensity-Based Visual Cryptography

With the increasing need of sensitive or secret data transmission through public network,security demands using cryptography and steganography are becoming a thirsty research area of last few years.These two techniques can be merged and provide better security which is nowadays extremely required.The proposed system provides a novel method of information security using the techniques of audio steganography combined with visual cryptography.In this system,we take a secret image and divide it into several subparts to make more than one incomprehensible sub-images using the method of visual cryptography.Each of the sub-images is then hidden within individual cover audio files using audio steganographic techniques.The cover audios are then sent to the required destinations where reverse steganography schemes are applied to them to get the incomprehensible component images back.At last,all the sub-images are superimposed to get the actual secret image.This method is very secure as it uses a two-step security mechanism to maintain secrecy.The possibility of interception is less in this technique because one must have each piece of correct sub-image to regenerate the actual secret image.Without superimposing every one of the sub-images meaningful secret images cannot be formed.Audio files are composed of densely packed bits.The high density of data in audio makes it hard for a listener to detect the manipulation due to the proposed time-domain audio steganographic method.

Smart Grid Wide-Area Transmission System Visualization

The installation of vast quantities of additional new sensing and communication equipment, in conjunction with building the computing infrastructure to store and manage data gathered by this equipment, has been the fi rst step in the creation of what is generically referred to as the "smart grid" for the electric transmission system. With this enormous capital investment in equipment having been made, attention is now focused on developing methods to analyze and visualize this large data set. The most direct use of this large set of new data will be in data visualization. This paper presents a survey of some visualization techniques that have been deployed by the electric power industry for visualizing data over the past several years. These techniques include pie charts, animation, contouring, time-varying graphs, geographic-based displays, image blending, and data aggregation techniques. The paper then emphasizes a newer concept of using word-sized graphics called sparklines as an extremely eff ective method of showing large amounts of timevarying data.

Integrating Audio-Visual Features and Text Information for Story Segmentation of News Video

Video data are composed of multimodal information streams including visual, auditory and textual streams, so an approach of story segmentation for news video using multimodal analysis is described in this paper. The proposed approach detects the topic-caption frames, and integrates them with silence clips detection results, as well as shot segmentation results to locate the news story boundaries. The integration of audio-visual features and text information overcomes the weakness of the approach using only image analysis techniques. On test data with 135 400 frames, when the boundaries between news stories are detected, the accuracy rate 85.8% and the recall rate 97.5% are obtained. The experimental results show the approach is valid and robust.

Transversal vibrations of double-plate systems

This paper presents an analytical and numerical analysis of free and forced transversal vibrations of an elastically connected double-plate system. Analytical solutions of a system of coupled partial differential equations, which describe corresponding dynamical free and forced processes, are obtained using Bernoulli＇s particular integral and Lagrange＇s method of variation constants. It is shown that one-mode vibrations correspond to two-frequency regime for free vibrations induced by initial conditions and to three-frequency regime for forced vibrations induced by one-frequency external excitation and corresponding initial conditions. The analytical solutions show that the elastic connec- tion between plates leads to the appearance of twofrequency regime of time function, which corresponds to one eigenamplitude function of one mode, and also that the time functions of different vibration modes are uncoupled, for each shape of vibrations. It has been proven that for both elastically connected plates, for every pair of m and n, two possibilities for appearance of the resonance dynamical states, as well as for appearance of the dynamical absorption, are present. Using the MathCad program, the corresponding visualizations of the characteristic forms of the plate middle surfaces through time are presented.

Expression of visual system homeobox 1 in human keratoconus

AIM: To investigate the expression of visual system homeobox 1(VSX1) and myofibroblast marker alpha smooth muscle actin(α-SMA) in keratoconus(KC). METHODS: Thirty corneal tissue were collected from KC patients after corneal transplantation and 15 normal donor corneas were obtained. All corneal tissues divided into 4 parts for different detections. Scanning electron microscopy was used to observe the ultrastructure of the specimens. VSX1 and α-SMA localization in cornea tissues was detected using immunofluorescence histochemistry. Reverse transcription-quantitative polymerase chain reaction(RT-qPCR) and Western blot were performed to analyze the expression level of VSX1 and α-SMA. RESULTS: Compared to normal cornea tissue, the collagen fibers in KC stroma were distortional and attenuated and keratocytes were abnormally changed. VSX1 and α-SMA located in the corneal stroma. The mRNA and protein expression level of VSX1 in KC were about 3 times as high as that of normal tissue(P<0.001). α-SMA was hardly expressed in the normal corneas, however, its expression in the KC was about 1.5 times higher than that of the normal corneas(P<0.0001). CONCLUSION: Compared with normal corneal the expression of VSX1 and α-SMA in KC both increased. VSX1 is related to the activation of keratocytes and involved in the pathogenesis of keratoconus.

Visualized User Interface for Decision Support System

To work efficiently with DSS, most users need assistance in representation conversion, i. e., translating the specific outcome from the DSS into the universal language of visual. In generally, it is much easier to understand the results from the DSS if they are translated into charts, maps, and other scientific displays, because visualization exploits human natural ability to recognize and understand visual pattern. In this paper we discuss the concept of visualization for DSS. AniGraftool, a software system, is introduced as an example of Visualized User Interface for DSS.

Functional outcome of the low vision aids for visual impairment secondary to central nervous system tumors in children

To assess functional outcomes of optical low vision aids(LVAs) for pediatric visual impairment due to central nervous system(CNS) tumors. A prospective case study was conducted on 15 children with history of CNS tumors with mean age of 10.47±1.85 y. Lighthouse distance, near visual acuity tests, cycloplegic refraction, reading speed measurement and visual field examination were done. Prescription of far and near LVAs followed by training sessions. LVPrasad-functional vision questionnaire was done to evaluate performance. Visual impairment was moderate(13.3%), severe(73.3%), profound(6.7%) and near blindness in 6.7%. Telescopes prescribed in 33.4%, video magnifier in 46.7%. Questionnaire scores were significantly improved for distant rather than near tasks(P≤0.05) after training. LVAs rehabilitation is an effective method of improving vision in pediatric visual defects secondary to CNS tumors.

Real-time Audio &Video Transmission System Based on Visible Light Communication

With the increasing popularity of solid sate lighting devices, Visible Light Communication (VLC) is globally recognized as an advanced and promising technology to realize short-range, high speed as well as large capacity wireless data transmission. In this paper, we propose a prototype of real-time audio and video broadcast system using inexpensive commercially available light emitting diode (LED) lamps. Experimental results show that real-time high quality audio and video with the maximum distance of 3 m can be achieved through proper layout of LED sources and improvement of concentration effects. Lighting model within room environment is designed and simulated which indicates close relationship between layout of light sources and distribution of illuminance.

Audio Highlight Detection Method for Cloud-Based Multimedia Service System

With the development of cloud-based data centers and multimedia technologies, cloud-based multimedia service systems have been paid more and more attention. Audio highlights detection plays an important role in the cloud-based multimedia service system. In this paper, we proposed a novel highlight detection method to extract the audio highlight effects for the cloud-based multimedia service system using the unsupervised approach. In the proposed method, we first extract the audio features for each audio document. Then the spectral clustering scheme was used to decompose the audio document into several audio effects. Then, we introduce the TF-IDF method to label the highlight effect. We design some experiments to evaluate the performance of the proposed method, and the experimental results show that our method can achieve satisfying results.

Audio-visual emotion recognition with multilayer boosted HMM

Emotion recognition has become an important task of modern human-computer interac- tion. A multilayer boosted HMM （ MBHMM ） classifier for automatic audio-visual emotion recognition is presented in this paper. A modified Baum-Welch algorithm is proposed for component HMM learn- ing and adaptive boosting （AdaBoost） is used to train ensemble classifiers for different layers （cues）. Except for the first layer, the initial weights of training samples in current layer are decided by recognition results of the ensemble classifier in the upper layer. Thus the training procedure using current cue can focus more on the difficult samples according to the previous cue. Our MBHMM clas- sifier is combined by these ensemble classifiers and takes advantage of the complementary informa- tion from multiple cues and modalities. Experimental results on audio-visual emotion data collected in Wizard of Oz scenarios and labeled under two types of emotion category sets demonstrate that our approach is effective and promising.