Robust speech recognition using speech enhancement

Automatic Speech Recognition (ASR) has matured into a technology which is becoming more common in our everyday lives, and is emerging as a necessity to minimise driver distraction when operating in-car systems such as navigation and infotainment. In “noise-free” environments, word recognition performance of these systems has been shown to approach 100%, however this performance degrades rapidly as the level of background noise is increased. Speech enhancement is a popular method for making ASR systems more ro- bust. Single-channel spectral subtraction was originally designed to improve hu- man speech intelligibility and many attempts have been made to optimise this algorithm in terms of signal-based metrics such as maximised Signal-to-Noise Ratio (SNR) or minimised speech distortion. Such metrics are used to assess en- hancement performance for intelligibility not speech recognition, therefore mak- ing them sub-optimal ASR applications. This research investigates two methods for closely coupling subtractive-type enhancement algorithms with ASR: (a) a computationally-efficient Mel-filterbank noise subtraction technique based on likelihood-maximisation (LIMA), and (b) in- troducing phase spectrum information to enable spectral subtraction in the com- plex frequency domain. Likelihood-maximisation uses gradient-descent to optimise parameters of the enhancement algorithm to best fit the acoustic speech model given a word se- quence known a priori. Whilst this technique is shown to improve the ASR word accuracy performance, it is also identified to be particularly sensitive to non-noise mismatches between the training and testing data. Phase information has long been ignored in spectral subtraction as it is deemed to have little effect on human intelligibility. In this work it is shown that phase information is important in obtaining highly accurate estimates of clean speech magnitudes which are typically used in ASR feature extraction. Phase Estimation via Delay Projection is proposed based on the stationarity of sinusoidal signals, and demonstrates the potential to produce improvements in ASR word accuracy in a wide range of SNR. Throughout the dissertation, consideration is given to practical implemen- tation in vehicular environments which resulted in two novel contributions – a LIMA framework which takes advantage of the grounding procedure common to speech dialogue systems, and a resource-saving formulation of frequency-domain spectral subtraction for realisation in field-programmable gate array hardware. The techniques proposed in this dissertation were evaluated using the Aus- tralian English In-Car Speech Corpus which was collected as part of this work. This database is the first of its kind within Australia and captures real in-car speech of 50 native Australian speakers in seven driving conditions common to Australian environments.

Creative Commons : the next generation : Creative Commons licence use five years on

Since its launch in 2001, the Creative Commons open content licensing initiative has received both praise and censure. While some have touted it as a major step towards removing the burdens copyright law imposes on creativity and innovation in the digital age, others have argued that it robs artists of their rightful income. This paper aims to provide a brief overview and analysis of the practical application of the Creative Commons licences five years after their launch. It looks at how the Creative Commons licences are being used and who is using them, and attempts to identify likely motivations for doing so. By identifying trends in how this licence use has changed over time, it also attempts to rebut arguments that Creative Commons is a movement of academics and hobbyists, and has no value for traditional organisations or working artists.

Asia and the Commons : case studies 2008

A collection of case studies of individuals and organisations utilising open models in the Asia Pacific and associated regions. The case studies represent activities in nine countries, broader regions such as the Arab nations, and global efforts towards sustainability and social justice, revealing creative ways of participating in the commons. Featured are remix artists, performers, open source software programmers, film makers, collecting institutions and publishing houses focused on democracy and change, who demonstrate a diverse set of motivations to engage with the shared ideals of openness and community collaboration.

Preliminary review of over-regulation in Australian financial services

This article examines a preliminary review and the limited evidence of over-regulation in Australian financial services. The 1997 Wallis Report and the CLERP 6 paper resulted in the amendments to Ch 7 of the Corporations Act 2001 (Cth) by the Financial Services Reform Act. Nearly a decade later the system based upon 'one-size fits all' dual track regime and a consistent licensing regime has greatly increased the costs of compliance. In the area of enforcement there has not been a dramatic change to the effective techniques applied by ASIC over other agencies such as APRA. In particular there are clear economic arguments, as well as international experiences which state that a single financial services regulator is more effective than the multi-layered approach adopted in Australia. Finally, in the superannuation area of financial services, which is worth A$800 billion there is unnecessary dual licensing and duplicated regulation with little evidence of any consumer-member benefit but at a much greater cost

Non-motorized public transport : a global review and analysis of trends and issues

Over recent decades there has been growing interest in the role of non-motorized modes in the overall transport system (especially walking and cycling for private purposes) and many government initiatives have been taken to encourage these active modes. However there has been relatively little research attention given to the paid form of non-motorized travel which can be called non-motorized public transport (NMPT). This involves cycle-powered vehicles which can carry several passengers (plus the driver) and a small amount of goods; and which provide flexible hail-and-ride services. Effectively they are non-motorized taxis. Common forms include cycle-rickshaw (Bangladesh, India), becak (Indonesia), cyclos (Vietnam, Cambodia), bicitaxi (Columbia, Cuba), velo-taxi (Germany, Netherland), and pedicabs (UK, Japan, USA). --------- The popularity of NMPT is widespread in developing countries, where it caters for a wide range of mobility needs. For instance in Dhaka, Bangladesh, rickshaws are the preferred mode for non-walk trips and have a higher mode share than cars or buses. Factors that underlie the continued existence and popularity of NMPT in many developing countries include positive contribution to social equity, micro-macro economic significance, employment creation, and suitability for narrow and crowded streets. Although top speeds are lower than motorized modes, NMPT is competitive and cost-effective for short distance door-to-door trips that make up the bulk of travel in many developing cities. In addition, NMPT is often the preferred mode for vulnerable groups such as females, children and elderly people. NMPT is more prominent in developing countries but its popularity and significance is also gradually increasing in several developed countries of Asia, Europe and parts of North America, where there is a trend for the NMPT usage pattern to broaden from tourism to public transport. This shift is due to a number of factors including the eco-sustainable nature of NMPT; its operating flexibility (such as in areas where motorized vehicle access is restricted or discouraged through pricing); and the dynamics that it adds to the urban fabric. Whereas NMPT may have been seen as a “dying” mode, in many cities it is maintaining or increasing its significance and with potential for further growth. --------- This paper will examine and analyze global trends in NMPT incorporating both developing and developed country contexts and issues such as usage patterns; NMPT policy and management practices; technological development; and operational integration of NMPT into the overall transport system. It will look at how NMPT policies, practices and usage have changed over time and the differing trends in developing and developed countries. In particular, it will use Dhaka, Bangladesh as a case study in recognition of its standing as the major NMPT city in the world. The aim is to highlight NMPT issues and trends and their significance for shaping future policy towards NMPT in developing and developed countries. The paper will be of interest to transport planners, traffic engineers, urban and regional planners, environmentalists, economists and policy makers.

Homophobia, heteronormativism and hegemonic masculinity : male same-sex intimate violence from the perspective of Brisbane service providers

Few studies have explored the problem of male same-sex intimate partner violence, especially in the context of Australia. Utilizing in-depth interviews with gay-friendly service providers in Brisbane, the research presented in this article sought to ascertain whether (a) intimate partner violence occurs in male same-sex intimate relationships, (b)if so, what form this violence takes,(c) what contextual triggers underpin this violence,(d) what barriers victims face in exiting abusive relationships and seeking support, and (e) what services are available and appropriate to the needs of men in violent intimate relationships with other men. Results suggest that the prevalence, types and contextual triggers of violence in male same-sex relationships parallel abuse in opposite-sex relationships. Heteronormativism, homophobia, and its close association with hegemonic masculinity, however, emerge as features unique to the male same-sex intimate partner violence experience.

High Court changes position on standard of care of learner drivers

Since the High Court decision of Cook v Cook (1986) 162 CLR 376, a person who voluntarily undertakes to instruct a learner driver of a motor vehicle is owed a lower standard of care than that owed to other road users. The standard of care was still expressed to be objective; however, it took into account the inexperience of the learner driver. Therefore, a person instructing a learner driver was owed a duty of care the standard being that of a reasonable learner driver. This ‘special relationship’ was said to exist because of the passenger’s knowledge of the driver’s inexperience and lack of skill. On 28 August 2008 the High Court handed down its decision in Imbree v McNeilly [2008] HCA 40, overruling Cook v Cook.

CC & Government Guide : Using Creative Commons 2.5 Australia Licences on Government Copyright Materials

This guide explains how copyright law applies to Australian government material, how copyright can be managed to facilitate beneficial open access practices by government, how CC licences can be used to achieve open access to government material, and provides practical step-by-step guidance for agencies and their officers on licensing and use of government copyright materials under CC 2.5 Australia licences.

Construction performance and organisational culture

The purpose of this paper is to summarize the outcomes of a detailed research study carried out as part of the fulfilment of a doctoral programme which examined the relationships between, and impacts of organisational culture on construction performance within a Hong Kong context. The research used a mixed methodology approach consisting of an organisational culture survey using an adapted validated and reliable measurement instrument (the Denison Organisational Culture Survey), mini-case studies in four Hong Kong construction companies and correlated the derived culture scores against performance scores measured by the Hong Kong Housing Department Performance Assessment Scoring System (PASS). The significance of the research was to advance knowledge of the importance of organisational culture strength as a performance driver in the construction industry and the further proof of the culture performance links using a set of measures of the latter which were not financially-based. The findings of the research make a contribution to theory by further validating the work by Denison (1990) and others, not only in that a successful link between organisational culture and performance was demonstrated, but it also identifies particular cultural factors in organisations that appear to be significantly responsible for achieving successful outcomes and reveals opportunities for further research into the organisational culture of construction companies Keywords: organisational culture, construction performance, business success.

Studying the effect of market orientation and organisational culture on a brand’s performance

It is now well accepted that effective implementation of market orientation leads to superior performance. This paper theorises that market orientation and an innovative culture enable organisations to achieve higher brand performance. To test this proposition data were gathered from a sample of firms across a range of industries. The results support the premise that market orientation and innovative cultures improve brand performance and that innovative culture influences market orientation. The results also indicate that innovative culture is the stronger driver of brand performance over market orientation.

Lip detection for audio-visual speech recognition in-car environment

Acoustically, car cabins are extremely noisy and as a consequence audio-only, in-car voice recognition systems perform poorly. As the visual modality is immune to acoustic noise, using the visual lip information from the driver is seen as a viable strategy in circumventing this problem by using audio visual automatic speech recognition (AVASR). However, implementing AVASR requires a system being able to accurately locate and track the drivers face and lip area in real-time. In this paper we present such an approach using the Viola-Jones algorithm. Using the AVICAR [1] in-car database, we show that the Viola- Jones approach is a suitable method of locating and tracking the driver’s lips despite the visual variability of illumination and head pose for audio-visual speech recognition system.

Health protection

Our understanding of how the environment can impact human health has evolved and expanded over the centuries, with concern and interest dating back to ancient times. For example, over 4000 years ago, a civilisation in northern India tried to protect the health of its citizens by constructing and positioning buildings according to strict building laws, by having bathrooms and drains, and by having paved streets with a sewerage system (Rosen 1993). In more recent times, the ‘industrial revolution’ played a dominant role in shaping the modern world, and with it the modern public health system. This era was signified by rapid progress in technology, the growth of transportation and the expansion of the market economy, which lead to the organisation of industry into a factory system. This meant that labour had to be brought to the factories and by the 1820s, poverty and social distress (including overcrowding and infrequent sewage and garbage disposal) was more widespread than ever. These circumstances, therefore, lead to the rise of the ‘sanitary revolution’ and the birth of modern public health (Rosen 1993). The sanitary revolution has also been described as constituting the beginning of the first wave of environmental concern, which continued until after World War 2 when major advances in engineering and chemistry substantially changed the face of industry, particularly the chemical sector. The second wave of environmental concern came in the mid to late 20th century and was dominated by the environmental or ecology movement. A landmark in this era was the 1962 publication of the book Silent Spring by Rachel Carson. This identified for the first time the dramatic effects on the ecosystem of the widespread use of the organochlorine pesticide, DDT. The third wave of environmental concern commenced in the 1980s and continues today. The accelerated rate of economic development, the substantial increase in the world population and the globalisation of trade have dramatically changed the production methods and demand for goods in both developed and developing countries. This has lead to the rise of ‘sustainable development’ as a key driver in environmental planning and economic development (Yassi et al 2001). The protection of health has, therefore, been a hallmark of human history and is the cornerstone of public health practice. This chapter introduces environmental health and how it is managed in Australia, including a discussion of the key generic management tools. A number of significant environmental health issues and how they are specifically managed are then discussed, and the chapter concludes by discussing sustainable development and its links with environmental health.

Driving performance on monotonous roads

Driving on motorways has largely been reduced to a lane-keeping task with cruise control. Rapidly, drivers are likely to get bored with such a task and take their attention away from the road. This is of concern in terms of road safety – particularly for professional drivers - since inattention has been identified as one of the main contributing factors to road crashes and is estimated to be involved in 20 to 30% of these crashes. Furthermore, drivers are not aware that their vigilance level has decreased and that their driving performance is impaired. Intelligent Transportation System (ITS) intervention can be used as a countermeasure against vigilance decrement. This paper aims to identify a variety of metrics impacted during monotonous driving - ranging from vehicle data to physiological variables - and relate them to two monotonous factors namely the monotony of the road design (straightness) and the monotony of the environment (landscape, signage, traffic). Data are collected in a driving simulator instrumented with an eye tracking system, a heart rate monitor and an electrodermal activity device (N=25 participants). The two monotonous factors are varied (high and low) leading to the use of four different driving scenarios (40 minutes each). We show with Generalised Linear Mixed Models that driver performance decreases faster when the road is monotonous. We also highlight that road monotony impairs a variety of driving performance and vigilance measures, ranging from speed, lateral position of the vehicle to physiological measurements such as heart rate variability, blink frequency and electrodermal activity. This study informs road designers of the importance of having a varied road environment. It also provides a range of metrics that can be used to detect in real-time the impairment of driving performance on monotonous roads. Such knowledge could result in the development of an in-vehicle device warning drivers at early signs of driving performance impairment on monotonous roads.

Global knowledge cultures

Knowledge is about cultural power. Considering that it is both resource and product within the brave new world of fast capitalism, this collection argues for knowledge cultures that are mutually engaged and hence more culturally inclusive and socially productive. Globalized intellectual property regimes, the privatization of information, and their counterpoint, the information and creative commons movements, constitute productive sites for the exploration of epistemologies that talk with each other rather than at and past each other. Global Knowledge Cultures provides a collection of accessible essays by some of the world’s leading legal scholars, new media analysts, techno activists, library professionals, educators and philosophers. Issues canvassed by the authors include the ownership of knowledge, open content licensing, knowledge policy, the common-wealth of learning, transnational cultural governance, and information futures. Together, they call for sustained intercultural dialogue for more ethical knowledge cultures within contexts of fast knowledge capitalism.

Safety implications of increased moped and scooter use in an Australian city

Drawing on two studies within a larger program of research into scooter and moped safety in Queensland, Australia, some key safety concerns specific to the use of these vehicles are discussed. A five phase observational study is used to identify distribution of powered two-wheeler (PTW) types in the city centre of Brisbane, Australia’s third largest city. Data were first collected in August 2008, and thereafter at six-monthly intervals. Stationary PTWs were directly observed in designated parking areas. Four focus groups involving 23 Brisbane riders were held in March 2009, aiming to explore perspectives on safety and transport planning in a semi-structured format. Information gathered in the focus groups informed development of a questionnaire targeting a larger sample of scooter and moped riders. The observations made to date indicate that 36% of all PTWs parked in Brisbane’s inner city are either mopeds or larger scooters, with the remaining 64% accounted for by motorcycles (n = 2037). These data suggest that mopeds and scooters are a significant transport mode in Brisbane, yet little is known about their safety relative to that of motorcycles. In focus groups, main motivating factors for scooter or moped use included parking availability, traffic congestion, cost, time-efficiency and enjoyment. Moped riders were generally younger and less experienced than other scooter riders, less likely to wear protective clothing, and less likely to have undertaken rider training. The focus groups have helped to identify some particular safety concerns regarding moped use in a jurisdiction requiring no motorcycle licence or rider training.