Syllable language models for Mandarin speech recognition: exploiting character language models.

Mandarin Chinese is based on characters which are syllabic in nature and morphological in meaning. All spoken languages have syllabiotactic rules which govern the construction of syllables and their allowed sequences. These constraints are not as restrictive as those learned from word sequences, but they can provide additional useful linguistic information. Hence, it is possible to improve speech recognition performance by appropriately combining these two types of constraints. For the Chinese language considered in this paper, character level language models (LMs) can be used as a first level approximation to allowed syllable sequences. To test this idea, word and character level n-gram LMs were trained on 2.8 billion words (equivalent to 4.3 billion characters) of texts from a wide collection of text sources. Both hypothesis and model based combination techniques were investigated to combine word and character level LMs. Significant character error rate reductions up to 7.3% relative were obtained on a state-of-the-art Mandarin Chinese broadcast audio recognition task using an adapted history dependent multi-level LM that performs a log-linearly combination of character and word level LMs. This supports the hypothesis that character or syllable sequence models are useful for improving Mandarin speech recognition performance.

Discriminative spoken language understanding using word confusion networks

Current commercial dialogue systems typically use hand-crafted grammars for Spoken Language Understanding (SLU) operating on the top one or two hypotheses output by the speech recogniser. These systems are expensive to develop and they suffer from significant degradation in performance when faced with recognition errors. This paper presents a robust method for SLU based on features extracted from the full posterior distribution of recognition hypotheses encoded in the form of word confusion networks. Following [1], the system uses SVM classifiers operating on n-gram features, trained on unaligned input/output pairs. Performance is evaluated on both an off-line corpus and on-line in a live user trial. It is shown that a statistical discriminative approach to SLU operating on the full posterior ASR output distribution can substantially improve performance both in terms of accuracy and overall dialogue reward. Furthermore, additional gains can be obtained by incorporating features from the previous system output. © 2012 IEEE.

A preliminary study of target multiplicity for ADSRs

The Accelerator Driven Subcritical Reactor (ADSR) concept is based on the coupling of a particle accelerator to a subcritical reactor core by means of a neutron spallation target interface. This paper investigates the benefits of multiple spallation targets in ADSRs. The motivation behind this is, firstly, to improve the overall reliability of the accelerator-reactor system, and, secondly, to evaluate other potential advantages such as lower beam power requirements. The results show that a system containing two or three spallation targets, coupled to independent accelerators, offers better neutronic performance. This is demonstrated through the increased effective multiplication factor (keff) in the two- and three-target configurations and a more uniform neutron flux distribution. A multiple-target ADSR also proves effective in mitigating the impact of frequent beam interruptions, a pressing issue that needs to be addressed for the ADSR concept to advance. Assuming no simultaneous beam shutdowns, the two- and three-target configurations reduce the risk of fuel cladding failure due to thermal cyclic fatigue. © 2013 Elsevier B.V. All rights reserved.

Language model cross adaptation for LVCSR system combination

State-of-the-art large vocabulary continuous speech recognition (LVCSR) systems often combine outputs from multiple sub-systems that may even be developed at different sites. Cross system adaptation, in which model adaptation is performed using the outputs from another sub-system, can be used as an alternative to hypothesis level combination schemes such as ROVER. Normally cross adaptation is only performed on the acoustic models. However, there are many other levels in LVCSR systems' modelling hierarchy where complimentary features may be exploited, for example, the sub-word and the word level, to further improve cross adaptation based system combination. It is thus interesting to also cross adapt language models (LMs) to capture these additional useful features. In this paper cross adaptation is applied to three forms of language models, a multi-level LM that models both syllable and word sequences, a word level neural network LM, and the linear combination of the two. Significant error rate reductions of 4.0-7.1% relative were obtained over ROVER and acoustic model only cross adaptation when combining a range of Chinese LVCSR sub-systems used in the 2010 and 2011 DARPA GALE evaluations. © 2012 Elsevier Ltd. All rights reserved.

Paraphrastic language models

In natural languages multiple word sequences can represent the same underlying meaning. Only modelling the observed surface word sequence can result in poor context coverage, for example, when using n-gram language models (LM). To handle this issue, this paper presents a novel form of language model, the paraphrastic LM. A phrase level transduction model that is statistically learned from standard text data is used to generate paraphrase variants. LM probabilities are then estimated by maximizing their marginal probability. Significant error rate reductions of 0.5%-0.6% absolute were obtained on a state-ofthe-art conversational telephone speech recognition task using a paraphrastic multi-level LM modelling both word and phrase sequences.

User target intention recognition from cursor position using kalman filter

This paper discusses user target intention recognition algorithms for pointing - clicking tasks to reduce users' pointing time and difficulty. Predicting targets by comparing the bearing angles to targets proposed as one of the first algorithms [1] is compared with a Kalman Filter prediction algorithm. Accuracy and sensitivity of prediction are used as performance criteria. The outcomes of a standard point and click experiment are used for performance comparison, collected from both able-bodied and impaired users. © 2013 Springer-Verlag Berlin Heidelberg.

Unsteady penetration of a target by a liquid jet.

It is widely acknowledged that ceramic armor experiences an unsteady penetration response: an impacting projectile may erode on the surface of a ceramic target without substantial penetration for a significant amount of time and then suddenly start to penetrate the target. Although known for more than four decades, this phenomenon, commonly referred to as dwell, remains largely unexplained. Here, we use scaled analog experiments with a low-speed water jet and a soft, translucent target material to investigate dwell. The transient target response, in terms of depth of penetration and impact force, is captured using a high-speed camera in combination with a piezoelectric force sensor. We observe the phenomenon of dwell using a soft (noncracking) target material. The results show that the penetration rate increases when the flow of the impacting water jet is reversed due to the deformation of the jet-target interface--this reversal is also associated with an increase in the force exerted by the jet on the target. Creep penetration experiments with a constant indentation force did not show an increase in the penetration rate, confirming that flow reversal is the cause of the unsteady penetration rate. Our results suggest that dwell can occur in a ductile noncracking target due to flow reversal. This phenomenon of flow reversal is rather widespread and present in a wide range of impact situations, including water-jet cutting, needleless injection, and deposit removal via a fluid jet.

Paraphrastic language models and combination with neural network language models

In natural languages multiple word sequences can represent the same underlying meaning. Only modelling the observed surface word sequence can result in poor context coverage, for example, when using n-gram language models (LM). To handle this issue, paraphrastic LMs were proposed in previous research and successfully applied to a US English conversational telephone speech transcription task. In order to exploit the complementary characteristics of paraphrastic LMs and neural network LMs (NNLM), the combination between the two is investigated in this paper. To investigate paraphrastic LMs' generalization ability to other languages, experiments are conducted on a Mandarin Chinese broadcast speech transcription task. Using a paraphrastic multi-level LM modelling both word and phrase sequences, significant error rate reductions of 0.9% absolute (9% relative) and 0.5% absolute (5% relative) were obtained over the baseline n-gram and NNLM systems respectively, after a combination with word and phrase level NNLMs. © 2013 IEEE.

Automatic construction and natural-language description of nonparametric regression models

Copyright © 2014, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. This paper presents the beginnings of an automatic statistician, focusing on regression problems. Our system explores an open-ended space of statistical models to discover a good explanation of a data set, and then produces a detailed report with figures and natural- language text. Our approach treats unknown regression functions non- parametrically using Gaussian processes, which has two important consequences. First, Gaussian processes can model functions in terms of high-level properties (e.g. smoothness, trends, periodicity, changepoints). Taken together with the compositional structure of our language of models this allows us to automatically describe functions in simple terms. Second, the use of flexible nonparametric models and a rich language for composing them in an open-ended manner also results in state- of-the-art extrapolation performance evaluated over 13 real time series data sets from various domains.

Calibrating the Gaussian multi-target tracking model

We present novel batch and online (sequential) versions of the expectation-maximisation (EM) algorithm for inferring the static parameters of a multiple target tracking (MTT) model. Online EM is of particular interest as it is a more practical method for long data sets since in batch EM, or a full Bayesian approach, a complete browse of the data is required between successive parameter updates. Online EM is also suited to MTT applications that demand real-time processing of the data. Performance is assessed in numerical examples using simulated data for various scenarios. For batch estimation our method significantly outperforms an existing gradient based maximum likelihood technique, which we show to be significantly biased. © 2014 Springer Science+Business Media New York.

Unified form language: A domain-specific language for weak formulations of partial differential equations

We present the Unified Form Language (UFL), which is a domain-specific language for representing weak formulations of partial differential equations with a view to numerical approximation. Features of UFL include support for variational forms and functionals, automatic differentiation of forms and expressions, arbitrary function space hierarchies formultifield problems, general differential operators and flexible tensor algebra. With these features, UFL has been used to effortlessly express finite element methods for complex systems of partial differential equations in near-mathematical notation, resulting in compact, intuitive and readable programs. We present in this work the language and its construction. An implementation of UFL is freely available as an open-source software library. The library generates abstract syntax tree representations of variational problems, which are used by other software libraries to generate concrete low-level implementations. Some application examples are presented and libraries that support UFL are highlighted. © 2014 ACM.

Structural and optical properties of ZnO thin films on glass substrate grown by laser-ablating Zn target in oxygen atmosphere

C-axis-orientated ZnO thin films were prepared on glass substrates by pulsed-laser deposition (PLD) technique in an oxygen-reactive atmosphere, using a metallic Zn target. The effects of growth condition such as laser energy and substrate temperature on the structural and optical properties of ZnO films had been investigated by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission spectra and room-temperature (RT) photoluminescence (PL) measurements. The results showed that the thickness, crystallite size, and compactness of ZnO films increased with the laser energy and substrate temperature. Both the absorption edges and the UV emission peaks of the films exhibited redshift, and UV emission intensity gradually increased as the laser energy and substrate temperature increased. From these results, it was concluded that crystalline quality of ZnO films was improved with increasing laser energy and substrate temperature. (c) 2007 Elsevier B.N. All rights reserved.

A novel vision chip for high-speed target tracking

This paper presents a novel vision chip for high-speed target tracking. Two concise algorithms for high-speed target tracking are developed. The algorithms include some basic operations that can be used to process the real-time image information during target tracking. The vision chip is implemented that is based on the algorithms and a row-parallel architecture. A prototype chip has 64 x 64 pixels is fabricated by 0.35 pm complementary metal-oxide-semiconductor transistor (CMOS) process with 4.5 x 2.5 mm(2) area. It operates at a rate of 1000 frames per second with 10 MHz chip main clock. The experiment results demonstrate that a high-speed target can be tracked in complex static background and a high-speed target among other high-speed objects can be tracked in clean background.