Unsupervised speaker adaptation for telephone call transcription

The use of the PC and Internet for placing telephone calls will present new opportunities to capture vast amounts of un-transcribed speech for a particular speaker. This paper investigates how to best exploit this data for speaker-dependent speech recognition. Supervised and unsupervised experiments in acoustic model and language model adaptation are presented. Using one hour of automatically transcribed speech per speaker with a word error rate of 36.0%, unsupervised adaptation resulted in an absolute gain of 6.3%, equivalent to 70% of the gain from the supervised case, with additional adaptation data likely to yield further improvements. LM adaptation experiments suggested that although there seems to be a small degree of speaker idiolect, adaptation to the speaker alone, without considering the topic of the conversation, is in itself unlikely to improve transcription accuracy.

Call for action : investigating the role of business process management in green IS

Sustainable practices are more than ever on the radar screen of organizations, triggered by a growing demand of the wider population towards approaches and practices that can be considered "green" or "sustainable". Our specific intent with this call for action is to immerse deeper into the role of business processes, and specifically the contributions that the management of these processes can play in leveraging the transformative power of information systems (IS) in order to create environmentally sustainable organizations. Our key premise is that business and information technology (IT) managers need to engage in a process-focused discussion to enable a common, comprehensive understanding of process, and the process-centered opportunities for making these processes, and ultimately the organization as a process-centric entity, "green". Based on a business process lifecycle model, we propose possible avenues for future research.

Applying six-sigma in the service industry: a review and case study in call center services

This paper presents an extensive review on the services, six-sigma, and application of six-sigma in services. In order to improve service quality focus on service process is necessary. Six-sigma is a philosophy which also concentrates on the improvement of process. So, six-sigma if properly applied can be useful for services. This study focuses on the application aspect of six-sigma to wider range of services. The wider applicability of six-sigma depends on identification of key performance indicators(KPIs) for different types of service processes. A case study is conducted in call center services to identify, analyze and compare critical to quality characteristics (CTQs) and KPIs with other types of services available in literature. This study will be helpful to both practitioners and researchers.

Timed and Probabilistic Automata for Automatic Animal Call Recognition

Automatic Call Recognition is vital for environmental monitoring. Patten recognition has been applied in automatic species recognition for years. However, few studies have applied formal syntactic methods to species call structure analysis. This paper introduces a novel method to adopt timed and probabilistic automata in automatic species recognition based upon acoustic components as the primitives. We demonstrate this through one kind of birds in Australia: Eastern Yellow Robin.

Detecting Symbian OS malware through static function call analysis

Smartphones become very critical part of our lives as they offer advanced capabilities with PC-like functionalities. They are getting widely deployed while not only being used for classical voice-centric communication. New smartphone malwares keep emerging where most of them still target Symbian OS. In the case of Symbian OS, application signing seemed to be an appropriate measure for slowing down malware appearance. Unfortunately, latest examples showed that signing can be bypassed resulting in new malware outbreak. In this paper, we present a novel approach to static malware detection in resource-limited mobile environments. This approach can be used to extend currently used third-party application signing mechanisms for increasing malware detection capabilities. In our work, we extract function calls from binaries in order to apply our clustering mechanism, called centroid. This method is capable of detecting unknown malwares. Our results are promising where the employed mechanism might find application at distribution channels, like online application stores. Additionally, it seems suitable for directly being used on smartphones for (pre-)checking installed applications.

Research needed on cultural capital, STEM endeavours and instructional innovations : a call for "collaborations between nations"

New Australian curriculum documents and government initiatives advocate the inclusion of Asian perspectives, which is highly relevant to the STEM fields. For Australia and other countries, STEM education is an opportunity to develop competencies towards employment in high-demand areas, yet the world’s knowledge of STEM is changing rapidly, requiring continuous analysis to meet market demands. This paper presents the need for “collaborations between nations” through research to advance each country’s STEM agenda towards further globalisation of education with the sharing of knowledge. Research is needed on views of what constitutes cultural capital for STEM, which also involves understanding past and current STEM endeavours occurring within various countries. Most importantly for STEM education is uncovering instructional innovations aligned with countries’ cultures and STEM endeavours. Research questions are provided in this paper to stimulate ideas for investigating in these fields. Economically, and as demonstrated recently by Greece and Spain, countries throughout the world can no longer operate independently for advancing standards of living. The world needs to recognise interdependence not only in trade and resources but also through the knowledge base that exists within countries. Learning together globally means transitioning from independence to interdependence in STEM education that will help each country meet global demands.

Reconceptualizing digital service quality : a call-to-action and research approach

The nature of services and service delivery has been changing rapidly since the 1980’s when many seminal papers in services research were published. Services are increasingly digital, or have a digital component. Further, a large and heterogeneous literature, with competing and overlapping definitions, many of which are dated and inappropriate to contemporary digital services offerings is impeding progress in digital services research. In this conceptual paper, we offer a critical review of some existing conceptualizations of services and digital services. We argue that an inductive approach to understanding cognition about digital services is required to develop a taxonomy of digital services and a new vocabulary. We argue that this is a pre-requisite to theorizing about digital services, including understanding quality drivers, value propositions, and quality determinants for different digital service types. We propose a research approach for reconceptualising digital services and service quality, and outline methodological approaches and outcomes.

Introduction : Call Centres, the Networked Economy and the Value Chain

In July 2006, members of Monash University’s Australian Centre for Research in Employment and Work (ACREW) joined with colleagues from King’s College London to hold a conference focused on ‘Socially responsive, socially responsible approaches to employment and work’ (DeCieri et al., 2006). One of the five conference streams was devoted to work and employment in call centres and was organized by the lead author of this introduction (Rainnie). This special edition of the JIR includes a number of articles from that stream.

Rethinking the value of simulation methods in the information systems research field : a call for reconstructing contribution for a broader audience

The impact of simulation methods for social research in the Information Systems (IS) research field remains low. A concern is our field is inadequately leveraging the unique strengths of simulation methods. Although this low impact is frequently attributed to methodological complexity, we offer an alternative explanation – the poor construction of research value. We argue a more intuitive value construction, better connected to the knowledge base, will facilitate increased value and broader appreciation. Meta-analysis of studies published in IS journals over the last decade evidences the low impact. To facilitate value construction, we synthesize four common types of simulation research contribution: Analyzer, Tester, Descriptor, and Theorizer. To illustrate, we employ the proposed typology to describe how each type of value is structured in simulation research and connect each type to instances from IS literature, thereby making these value types and their construction visible and readily accessible to the general IS community.

Human resource management of Indian call centre representatives

This study examines how call centres adopt different types of human resource practices (involvement and control oriented) to manage frontline employees in Indian call centres. Data were collected from 250 call centre representatives to test the research hypotheses. The research model was analyzed using Mplus software. Findings showed that involvement and control oriented human resource practices resulted in more employee exhaustion and disengagement. Involvement oriented HRM had a positive impact on job satisfaction as well as, a positive relationship between employee exhaustion and disengagement. The findings suggest that, while involvement oriented HRM enhances job satisfaction, its implementation comes with a cost, that is, an increase in employee exhaustion and disengagement at work.

Acoustic feature extraction using perceptual wavelet packet decomposition for frog call classification

Frog protection has become increasingly essential due to the rapid decline of its biodiversity. Therefore, it is valuable to develop new methods for studying this biodiversity. In this paper, a novel feature extraction method is proposed based on perceptual wavelet packet decomposition for classifying frog calls in noisy environments. Pre-processing and syllable segmentation are first applied to the frog call. Then, a spectral peak track is extracted from each syllable if possible. Track duration, dominant frequency and oscillation rate are directly extracted from the track. With k-means clustering algorithm, the calculated dominant frequency of all frog species is clustered into k parts, which produce a frequency scale for wavelet packet decomposition. Based on the adaptive frequency scale, wavelet packet decomposition is applied to the frog calls. Using the wavelet packet decomposition coefficients, a new feature set named perceptual wavelet packet decomposition sub-band cepstral coefficients is extracted. Finally, a k-nearest neighbour (k-NN) classifier is used for the classification. The experiment results show that the proposed features can achieve an average classification accuracy of 97.45% which outperforms syllable features (86.87%) and Mel-frequency cepstral coefficients (MFCCs) feature (90.80%).

Application of image processing techniques for frog call classification

Frogs have received increasing attention due to their effectiveness for indicating the environment change. Therefore, it is important to monitor and assess frogs. With the development of sensor techniques, large volumes of audio data (including frog calls) have been collected and need to be analysed. After transforming the audio data into its spectrogram representation using short-time Fourier transform, the visual inspection of this representation motivates us to use image processing techniques for analysing audio data. Applying acoustic event detection (AED) method to spectrograms, acoustic events are firstly detected from which ridges are extracted. Three feature sets, Mel-frequency cepstral coefficients (MFCCs), AED feature set and ridge feature set, are then used for frog call classification with a support vector machine classifier. Fifteen frog species widely spread in Queensland, Australia, are selected to evaluate the proposed method. The experimental results show that ridge feature set can achieve an average classification accuracy of 74.73% which outperforms the MFCCs (38.99%) and AED feature set (67.78%).