Continuous asr for flexible incremental dialogue

Spoken dialogue systems provide a convenient way for users to interact with a machine using only speech. However, they often rely on a rigid turn taking regime in which a voice activity detection (VAD) module is used to determine when the user is speaking and decide when is an appropriate time for the system to respond. This paper investigates replacing the VAD and discrete utterance recogniser of a conventional turn-taking system with a continuously operating recogniser that is always listening, and using the recogniser 1-best path to guide turn taking. In this way, a flexible framework for incremental dialogue management is possible. Experimental results show that it is possible to remove the VAD component and successfully use the recogniser best path to identify user speech, with more robustness to noise, potentially smaller latency times, and a reduction in overall recognition error rate compared to using the conventional approach. © 2013 IEEE.

On-line policy optimisation of Bayesian spoken dialogue systems via human interaction

A partially observable Markov decision process has been proposed as a dialogue model that enables robustness to speech recognition errors and automatic policy optimisation using reinforcement learning (RL). However, conventional RL algorithms require a very large number of dialogues, necessitating a user simulator. Recently, Gaussian processes have been shown to substantially speed up the optimisation, making it possible to learn directly from interaction with human users. However, early studies have been limited to very low dimensional spaces and the learning has exhibited convergence problems. Here we investigate learning from human interaction using the Bayesian Update of Dialogue State system. This dynamic Bayesian network based system has an optimisation space covering more than one hundred features, allowing a wide range of behaviours to be learned. Using an improved policy model and a more robust reward function, we show that stable learning can be achieved that significantly outperforms a simulator trained policy. © 2013 IEEE.

Mired in the meaning: Sustainability in consultancy

This paper explores sustainability as it relates to the services, operations and management of engineering and design (E&D) consulting firms, revealing how sustainability has been adopted by the engineering consulting industry as a key influence on its business. The motivations behind sustainability as a business priority for E&D consulting do not appear to reflect a corporate belief or endorsement of sustainability principles, but rather a reflection of market, employee and government drivers. Through interview-based reporting, the implications of these motivations are the focus of this paper through an exploration of the meaning, understanding and operationalising of sustainability in engineering design, assessment of design and in-firm management and strategy.

Magnetic characterisation of large grain, bulk Y-Ba-Cu-O superconductor-soft ferromagnetic alloy hybrid structures

Large grain, bulk Y-Ba-Cu-O (YBCO) high temperature superconductors (HTS) have significant potential for use in a variety of practical applications that incorporate powerful quasi-permanent magnets. In the present work, we investigate how the trapped field of such magnets can be improved by combining bulk YBCO with a soft FeNi, ferromagnetic alloy. This involves machining the alloy into components of various shapes, such as cylinders and rings, which are attached subsequently to the top surface of a solid, bulk HTS cylinder. The effect of these modifications on the magnetic hysteresis curve and trapped field of the bulk superconductor at 77 K are then studied using pick-up coil and Hall probe measurements. The experimental data are compared to finite element modelling of the magnetic flux distribution using Campbell's algorithm. Initially we establish the validity of the technique involving pick-up coils wrapped around the bulk superconductor to obtain its magnetic hysteresis curve in a non-destructive way and highlight the difference between the measured signal and the true magnetization of the sample. We then consider the properties of hybrid ferromagnet/superconductor (F/S) structures. Hall probe measurements, together with the results of the model, establish that flux lines curve outwards through the ferromagnet, which acts, effectively, like a magnetic short circuit. Magnetic hysteresis curves show that the effects of the superconductor and the ferromagnet simply add when the ferromagnet is saturated fully by the applied field. The trapped field of the hybrid structure is always larger than that of the superconductor alone below this saturation level, and especially when the applied field is removed. The results of the study show further that the beneficial effects on the trapped field are enhanced when the ferromagnet covers the entire surface of the superconductor for different ferromagnetic components of various shapes and fixed volume. © 2014 Elsevier B.V. All rights reserved.