Unsupervised training for mandarin broadcast news and conversation transcription

A significant cost in obtaining acoustic training data is the generation of accurate transcriptions. For some sources close-caption data is available. This allows the use of lightly-supervised training techniques. However, for some sources and languages close-caption is not available. In these cases unsupervised training techniques must be used. This paper examines the use of unsupervised techniques for discriminative training. In unsupervised training automatic transcriptions from a recognition system are used for training. As these transcriptions may be errorful data selection may be useful. Two forms of selection are described, one to remove non-target language shows, the other to remove segments with low confidence. Experiments were carried out on a Mandarin transcriptions task. Two types of test data were considered, Broadcast News (BN) and Broadcast Conversations (BC). Results show that the gains from unsupervised discriminative training are highly dependent on the accuracy of the automatic transcriptions. © 2007 IEEE.

Evaluation of manufacturing techniques for a minifluidics demonstrator system

A multi-disciplinary team based at Heriot-Watt University and other Universities has been set up to tackle the design and manufacturing of lab-on-a-chip for industries as one of the demonstrators of the EPSRC Grand Challenge project "3D-Mintegration". The team focuses on the analysis of foetal genetic material extracted from maternal blood as a smart alternative to invasive prenatal testing such as amniocentesis. The first module of the microsystem envisaged achieves a separation of blood cells from plasma. This system permits the testing of different manufacturing techniques.

Evaluation of termination techniques for 4H-SiC PiN diodes and trench JFETs

An investigation concerning suitable termination techniques for 4H-SiC trench JFETs is presented. Field plates, p+ floating rings and junction termination extension techniques are used to terminate 1.2kV class PiN diodes. The fabricated PiN diodes evaluated here have a similar design to trench JFETs. Therefore, the conclusions for PiN diodes can be applied to JFET structures as well. Numerical simulations are also used to illustrate the effect of the terminations on the diodes' blocking mode behaviour.

Comparison of relative flexoelectric coefficients measured by different techniques

While it is well known that it is possible to determine the effective flexoelectric coefficient of nematic liquid crystals using hybrid cells [1], this technique can be difficult due to the necessity of using a D.C. field. We have used a second method[2], requiring an A.C. field, to determine this parameter and here we compare the two techniques. The A.C. method employs the linear flexoelectrically induced linear electro-optic switching mechanism observed in chiral nematics. In order to use this second technique a chiral nematic phase is induced in an achiral nematic by the addition of a small amount of chiral additive (∼3% concentration w/w) to give helix pitch lengths of typically 0.5-1.0 μm. We note that the two methods can be used interchangeably, since they produce similar results, and we conclude with a discussion of their relative merits.

Language model combination and adaptation using weighted finite state transducers

In speech recognition systems language model (LMs) are often constructed by training and combining multiple n-gram models. They can be either used to represent different genres or tasks found in diverse text sources, or capture stochastic properties of different linguistic symbol sequences, for example, syllables and words. Unsupervised LM adaptation may also be used to further improve robustness to varying styles or tasks. When using these techniques, extensive software changes are often required. In this paper an alternative and more general approach based on weighted finite state transducers (WFSTs) is investigated for LM combination and adaptation. As it is entirely based on well-defined WFST operations, minimum change to decoding tools is needed. A wide range of LM combination configurations can be flexibly supported. An efficient on-the-fly WFST decoding algorithm is also proposed. Significant error rate gains of 7.3% relative were obtained on a state-of-the-art broadcast audio recognition task using a history dependently adapted multi-level LM modelling both syllable and word sequences. ©2010 IEEE.

Parasitic feedthrough cancellation techniques for enhanced electrical characterization of electrostatic microresonators

Capacitive parasitic feedthrough is an impediment that is inherent to all electrically interfaced micron scale resonant devices, resulting in increased challenges to their integration in more complex circuits, particularly as devices are scaled to operate at higher frequencies for RF applications. In this paper, a technique to cancel the undesirable effects of capacitive feedthrough that was previously proposed is here developed for an on-chip implementation. The method reported in this paper benefits from the simplicity of its implementation, and its effectiveness is demonstrated in this paper. This technique is demonstrated for two disk-plate resonators that have been excited in the wine glass mode at 5.4 MHz, though applicable to almost any electrically interfaced resonator. Measurements of the electrical transmission from these resonators show that the magnitude of the frequency response of the system is enhanced by up to 19 dB, while the phase is found to shift through a full 180° about the resonant frequency. This method is proposed as a useful addition to other techniques for enhancing the measured response of electrostatic micromechanical resonators. © 2009 Elsevier B.V. All rights reserved.