Joint decoding of multiple speech patterns for robust speech recognition

We are addressing a new problem of improving automatic speech recognition performance, given multiple utterances of patterns from the same class. We have formulated the problem of jointly decoding K multiple patterns given a single Hidden Markov Model. It is shown that such a solution is possible by aligning the K patterns using the proposed Multi Pattern Dynamic Time Warping algorithm followed by the Constrained Multi Pattern Viterbi Algorithm The new formulation is tested in the context of speaker independent isolated word recognition for both clean and noisy patterns. When 10 percent of speech is affected by a burst noise at -5 dB Signal to Noise Ratio (local), it is shown that joint decoding using only two noisy patterns reduces the noisy speech recognition error rate to about 51 percent, when compared to the single pattern decoding using the Viterbi Algorithm. In contrast a simple maximization of individual pattern likelihoods, provides only about 7 percent reduction in error rate.

Understanding urban rail travel for improved patronage forecasting

Improved forecasting of urban rail patronage is essential for effective policy development and efficient planning for new rail infrastructure. Past modelling and forecasting of urban rail patronage has been based on legacy modelling approaches and often conducted at the general level of public transport demand, rather than being specific to urban rail. This project canvassed current Australian practice and international best practice to develop and estimate time series and cross-sectional models of rail patronage for Australian mainland state capital cities. This involved the implementation of a large online survey of rail riders and non-riders for each of the state capital cities, thereby resulting in a comprehensive database of respondent socio-economic profiles, travel experience, attitudes to rail and other modes of travel, together with stated preference responses to a wide range of urban travel scenarios. Estimation of the models provided a demonstration of their ability to provide information on the major influences on the urban rail travel decision. Rail fares, congestion and rail service supply all have a strong influence on rail patronage, while a number of less significant factors such as fuel price and access to a motor vehicle are also influential. Of note, too, is the relative homogeneity of rail user profiles across the state capitals. Rail users tended to have higher incomes and education levels. They are also younger and more likely to be in full-time employment than non-rail users. The project analysis reported here represents only a small proportion of what could be accomplished utilising the survey database. More comprehensive investigation was beyond the scope of the project and has been left for future work.

Cultures of sharing in 3D printing: What can we learn from the licence choices of Thingiverse users?

This article contributes to the discussion by analysing how users of the leading online 3D printing design repository Thingiverse manage their intellectual property (IP). 3D printing represents a fruitful case study for exploring the relationship between IP norms and practitioner culture. Although additive manufacturing technology has existed for decades, 3D printing is on the cusp of a breakout into the technological mainstream – hardware prices are falling; designs are circulating widely; consumer-friendly platforms are multiplying; and technological literacy is rising. Analysing metadata from more than 68,000 Thingiverse design files collected from the site, we examine the licensing choices made by users and explore the way this shapes the sharing practices of the site’s users. We also consider how these choices and practices connect with wider attitudes towards sharing and intellectual property in 3D printing communities. A particular focus of the article is how Thingiverse structures its regulatory framework to avoid IP liability, and the extent to which this may have a bearing on users’ conduct. The paper has three sections. First, we will offer a description of Thingiverse and how it operates in the 3D printing ecosystem, noting the legal issues that have arisen regarding Thingiverse’s Terms of Use and its allocation of intellectual property rights. Different types of Thingiverse licences will be detailed and explained. Second, the empirical metadata we have collected from Thingiverse will be presented, including the methods used to obtain this information. Third, we will present findings from this data on licence choice and the public availability of user designs. Fourth, we will look at the implications of these findings and our conclusions regarding the particular kind of sharing ethic that is present in Thingiverse; we also consider the “closed” aspects of this community and what this means for current debates about “open” innovation.

Understanding the chemistry of atmospheric aerosol particle formation via observations of physical properties

It is widely accepted that the global climate is heating up due to human activities, such as burning of fossil fuels. Therefore we find ourselves forced to make decisions on what measures, if any, need to be taken to decrease our warming effect on the planet before any irrevocable damage occurs. Research is being conducted in a variety of fields to better understand all relevant processes governing Earth s climate, and to assess the relative roles of anthropogenic and biogenic emissions into the atmosphere. One of the least well quantified problems is the impact of small aerosol particles (both of anthropogenic and biogenic origin) on climate, through reflecting solar radiation and their ability to act as condensation nuclei for cloud droplets. In this thesis, the compounds driving the biogenic formation of new particles in the atmosphere have been examined through detailed measurements. As directly measuring the composition of these newly formed particles is extremely difficult, the approach was to indirectly study their different characteristics by measuring the hygroscopicity (water uptake) and volatility (evaporation) of particles between 10 and 50 nm. To study the first steps of the formation process in the sub-3 nm range, the nucleation of gaseous precursors to small clusters, the chemical composition of ambient naturally charged ions were measured. The ion measurements were performed with a newly developed mass spectrometer, which was first characterized in the laboratory before being deployed at a boreal forest measurement site. It was also successfully compared to similar, low-resolution instruments. The ambient measurements showed that sulfuric acid clusters dominate the negative ion spectrum during new particle formation events. Sulfuric acid/ammonia clusters were detected in ambient air for the first time in this work. Even though sulfuric acid is believed to be the most important gas phase precursor driving the initial cluster formation, measurements of the hygroscopicity and volatility of growing 10-50 nm particles in Hyytiälä showed an increasing role of organic vapors of a variety of oxidation levels. This work has provided additional insights into the compounds participating both in the initial formation and subsequent growth of atmospheric new aerosol particles. It will hopefully prove an important step in understanding atmospheric gas-to-particle conversion, which, by influencing cloud properties, can have important climate impacts. All available knowledge needs to be constantly updated, summarized, and brought to the attention of our decision-makers. Only by increasing our understanding of all the relevant processes can we build reliable models to predict the long-term effects of decisions made today.

A Historical Approach to Children s Physics Education : Modelling of DC-circuit Phenomena in a Small Group

This three-phase design research describes the modelling processes for DC-circuit phenomena. The first phase presents an analysis of the development of the DC-circuit historical models in the context of constructing Volta s pile at the turn of the 18th century. The second phase involves the designing of a teaching experiment for comprehensive school third graders. Among other considerations, the design work utilises the results of the first phase and research literature of pupils mental models for DC-circuit phenomena. The third phase of the research was concerned with the realisation of the planned teaching experiment. The aim of this phase was to study the development of the external representations of DC-circuit phenomena in a small group of third graders. The aim of the study has been to search for new ways to guide pupils to learn DC-circuit phenomena while emphasing understanding at the qualitative level. Thus, electricity, which has been perceived as a difficult and abstract subject, could be learnt more comprehensively. Especially, the research of younger pupils learning of electricity concepts has not been of great interest at the international level, although DC-circuit phenomena are also taught in the lower classes of comprehensive schools. The results of this study are important, because there has tended to be more teaching of natural sciences in the lower classes of comprehensive schools, and attempts are being made to develop this trend in Finland. In the theoretical part of the research an Experimental-centred representation approach, which emphasises the role of experimentalism in the development of pupil s representations, is created. According to this approach learning at the qualitative level consists of empirical operations like experimenting, observations, perception, and prequantification of nature phenomena, and modelling operations like explaining and reasoning. Besides planning teaching, the new approach can be used as an analysis tool in describing both historical modelling and the development of pupils representations. In the first phase of the study, the research question was: How did the historical models of DC-circuit phenomena develop in Volta s time? The analysis uncovered three qualitative historical models associated with the historical concept formation process. The models include conceptions of the electric circuit as a scene in the DC-circuit phenomena, the comparative electric-current phenomenon as a cause of different observable effect phenomena, and the strength of the battery as a cause of the electric-current phenomenon. These models describe the concept formation process and its phases in Volta s time. The models are portrayed in the analysis using fragments of the models, where observation-based fragments and theoretical fragements are distinguished from each other. The results emphasise the significance of the qualitative concept formation and the meaning of language in the historical modelling of DC-circuit phenomena. For this reason these viewpoints are stressed in planning the teaching experiment in the second phase of the research. In addition, the design process utilised the experimentation behind the historical models of DC-circuit phenomena In the third phase of the study the research question is as follows: How will the small group s external representations of DC-circuit phenomena develop during the teaching experiment? The main question is divided into the following two sub questions: What kind of talk exists in the small group s learning? What kinds of external representations for DC-circuit phenomena exist in the small group discourse during the teaching experiment? The analysis revealed that the teaching experiment of the small group succeeded in its aim to activate talk in the small group. The designed connection cards proved especially successful in activating talk. The connection cards are cards that represent the components of the electric circuit. In the teaching experiment the pupils constructed different connections with the connection cards and discussed, what kinds of DC-circuit phenomena would take place in the corresponding real connections. The talk of the small group was analysed by comparing two situations, firstly, when the small group discussed using connections made with the connection cards and secondly with the same connections using real components. According to the results the talk of the small group included more higher-order thinking when using the connection cards than with similar real components. In order to answer the second sub question concerning the small group s external representations that appeared in the talk during the teaching experiment; student talk was visualised by the fragment maps which incorporate the electric circuit, the electric current and the source voltage. The fragment maps represent the gradual development of the external representations of DC-circuit phenomena in the small group during the teaching experiment. The results of the study challenge the results of previous research into the abstractness and difficulty of electricity concepts. According to this research, the external representations of DC-circuit phenomena clearly developed in the small group of third graders. Furthermore, the fragment maps uncover that although the theoretical explanations of DC-circuit phenomena, which have been obtained as results of typical mental model studies, remain undeveloped, learning at the qualitative level of understanding does take place.

The Perception of Light: Understanding Architectural Lighting Design

Light is essential to life and vision; without light, nothing exists. It plays a pivotal role in the world of architectural design and is used to generate all manner of perceptions that enhance the designed environment experience. But what are the fundamental elements that designers rely upon to generate light enhanced experiences? How are people’s perceptions influenced by designed light schemas? In this book Dr. Marisha McAuliffe highlights the relationship that exists between light source and surface and how both create quality of effect in the built environment. Concepts relating to architectural lighting design history, theories, research, and generation of lighting design schemes to create optimal experiences in architecture, interior architecture and design are all explored in detail. This book is essential reading for both the student and the professional working in architectural lighting, particularly in terms of qualitative perception oriented lighting design

Understanding the folding process of synthetic polymers by small-molecule folding agents

Two acceptor containing polyimides PDI and NDI carrying pyromellitic diimide units and 1,4,5,8-naphthalene tetracarboxy diimide units, respectively, along with hexa(oxyethylene) (EO6) segments as linkers, were prepared from the corresponding dianhydrides and diamines. These polyimides were made to fold by interaction with specifically designed folding agents containing a dialkoxynaphtha-lene (DAN) donor linked to a carboxylic acid group. The alkali-metal counter-ion of the donor carboxylic acid upon complexation with the EO6 segment brings the DAN unit in the right location to induce a charge-transfer complex formation with acceptor units in the polymer backbone. This two-point interaction between the folding agent and the polymer backbone leads to a folding of the polymer chain, which was readily monitored by NMR titrations. The effect of various parameters, such as structures of the folding agent and polymer, and the solvent composition, on the folding propensities of the polymer was studied.

Understanding the Knowledge Needs of Designers During Design Process in Industry

Product success is substantially influenced by satisfaction of knowledge needs of designers, and many tools and methods have been proposed to support these needs. However, adoption of these methods in industry is minimal. This may be due to an inadequate understanding of the knowledge needs of designers in industry. This research attempts to develop a better understanding of these needs by undertaking descriptive studies in an industry. We propose a taxonomy of knowledge, and evaluate this by analyzing the questions asked by the designers involved in the study during their interactions. Using the taxonomy, we converted the questions asked into a generic form. The generic questions provide an understanding about what knowledge must be captured during design, and what its structure should be.

Understanding Membranes through the Molecular Design of Lipids

Lipids are amphiphilic molecules that are composed of hydrophilic and hydrophobic regions. A typical membranous aggregate (vesicles, water-filled lipid nanospheres) is formed upon the self-organization of lipids in water from a diverse collection of amphiphiles producing a dynamic supramolecular structure that shows phase behavior and ordering as required for specific biological functions. The determination of various physical properties of lipid aggregates is the key to determining structure-function relationships. Over the years, we have designed and synthesized a wide variety of lipid molecular systems for the investigation of their membrane-forming properties and have used them for purposes such as gene delivery and enzyme activation. In this feature article, we focus on our work on various types of lipids including ion-paired amphiphiles, cholesterol-based lipids, aromatic lipids, macrocyclic lipids containing disulfide tethers; cationic dimeric lipids, and so forth. The emphasis is oil experimental design and bottom-line conclusions.

Understanding data flow and security requirements in wireless Body Area Networks for healthcare

The Body Area Network (BAN) is an emerging technology that focuses on monitoring physiological data in, on and around the human body. BAN technology permits wearable and implanted sensors to collect vital data about the human body and transmit it to other nodes via low-energy communication. In this paper, we investigate interactions in terms of data flows between parties involved in BANs under four different scenarios targeting outdoor and indoor medical environments: hospital, home, emergency and open areas. Based on these scenarios, we identify data flow requirements between BAN elements such as sensors and control units (CUs) and parties involved in BANs such as the patient, doctors, nurses and relatives. Identified requirements are used to generate BAN data flow models. Petri Nets (PNs) are used as the formal modelling language. We check the validity of the models and compare them with the existing related work. Finally, using the models, we identify communication and security requirements based on the most common active and passive attack scenarios.

A model based factorization approach for dense 3D recovery from monocular video

Feature track matrix factorization based methods have been attractive solutions to the Structure-front-motion (Sfnl) problem. Group motion of the feature points is analyzed to get the 3D information. It is well known that the factorization formulations give rise to rank deficient system of equations. Even when enough constraints exist, the extracted models are sparse due the unavailability of pixel level tracks. Pixel level tracking of 3D surfaces is a difficult problem, particularly when the surface has very little texture as in a human face. Only sparsely located feature points can be tracked and tracking error arc inevitable along rotating lose texture surfaces. However, the 3D models of an object class lie in a subspace of the set of all possible 3D models. We propose a novel solution to the Structure-from-motion problem which utilizes the high-resolution 3D obtained from range scanner to compute a basis for this desired subspace. Adding subspace constraints during factorization also facilitates removal of tracking noise which causes distortions outside the subspace. We demonstrate the effectiveness of our formulation by extracting dense 3D structure of a human face and comparing it with a well known Structure-front-motion algorithm due to Brand.

Tytöt, pojat ja sukupuolinen häirintä

In this research I ask what is interpreted as sex-based harassment by 15-16-year old girls and boys. By sex-based harassment I refer to one-sided, unwanted attention that is based on gender and that makes the target feel embarrassed, frightened, hurt or angry. My focus is not on the most overt cases of harassment but rather on everyday encounters. While young people differentiate between harassing and non-harassing attention, at the same time they define, assign value to and construct differences and power relations on the basis of gender, age and ethnicity, for example. My main data consists of essays (N 104, 54 girls, 54 boys) and thematic interviews (N 14; 20 girls, 3 boys) of ninth graders of a secondary school in Helsinki. In the essays and interviews, students construe the border between pleasant and unpleasant, tolerable and intolerable attention as clear in principle, but, they suggest that in practice this border is ambivalent, negotiable and contextual. The interpretations of incidents are justified by referring to features of the target, the scene or the perpetrator. Targets of harassment are most often construed as being girls who are characterized as thin-skinned, but at the same time they are expected to be understanding toward any sex-based attention they may get, particularly when it is not physical. On the other hand, girls are regarded as equal and even active participants in incidents of harassment. Such statements include considerations of how girls either reject or invite particular kinds of attention by their actions and outward appearance. Forms of harassment, ways of understanding it as well as overcoming it vary according to spatial context. By situating incidents in different spaces and places, young people contrast their experiences with ordinary and predictable non-harassment that takes place e.g. in discos and unusual and unexpected harassment that takes place e.g. in the city streets in the daytime. The behaviour of boys harassing a girls is naturalized by appealing to young masculinity and the childishness but also strong sexual drive which is seen as characteristic of teenage boys. On the other hand, sex-based harassment is racialized and pathologized in ways that separate the phenomenon from young, Finnish, normal masculinity. Both the material experiences of the young people and the definitions of the parties involved in harassing incidents are gendered. Girls encounter and deal with sexualized commenting and unwanted approaches much more often and in a more intensive way than boys. Furthermore, there is a vast cultural repertoire of acceptable accounts that can be mobilised in order to excuse male harassers, to critically evaluate the appearance or action of the female targets and to divide the responsibility between the female target and the male perpetrator.

Plumbing systems of shallow level intrusive complexes

We have come a long way from simple straw and balloon models of magma plumbing systems to a more detailed picture of shallow level intrusive complexes. In this chapter, the sub-volcanic plumbing system is considered in terms of how we can define the types and styles of magma networks from the deep to the shallow subsurface. We look at the plumbing system from large igneous provinces, through rifted systems to polygenetic volcanoes, with a view to characterising some of the key conceptual models. There is a focus on how ancient magmatic centres can help us better understand magmatic plumbing. New innovative ways to consider and quantify magma plumbing are also highlighted including 3D seismic, and using the crystal cargo to help fingerprint key magma plumbing events. Conclusions are drawn to our understanding of the 3D plumbing system and how these recent advances can be helpful when exploring the other chapters of this contribution.

Handling sectional views in volume-based approach to automatically construct 3D solid from 2D views

This paper presents an algorithm for solid model reconstruction from 2D sectional views based on volume-based approach. None of the existing work in automatic reconstruction from 2D orthographic views have addressed sectional views in detail. It is believed that the volume-based approach is better suited to handle different types of sectional views. The volume-based approach constructs the 3D solid by a boolean combination of elementary solids. The elementary solids are formed by sweep operation on loops identified in the input views. The only adjustment to be made for the presence of sectional views is in the identification of loops that would form the elemental solids. In the algorithm, the conventions of engineering drawing for sectional views, are used to identify the loops correctly. The algorithm is simple and intuitive in nature. Results have been obtained for full sections, offset sections and half sections. Future work will address other types of sectional views such as removed and revolved sections and broken-out sections. (C) 2004 Elsevier Ltd. All rights reserved.