Improving speech transcription for Mandarin-english translation

This paper describes the development of the CU-HTK Mandarin Speech-To-Text (STT) system and assesses its performance as part of a transcription-translation pipeline which converts broadcast Mandarin audio into English text. Recent improvements to the STT system are described and these give Character Error Rate (CER) gains of 14.3% absolute for a Broadcast Conversation (BC) task and 5.1% absolute for a Broadcast News (BN) task. The output of these STT systems is then post-processed, so that it consists of sentence-like segments, and translated into English text using a Statistical Machine Translation (SMT) system. The performance of the transcription-translation pipeline is evaluated using the Translation Edit Rate (TER) and BLEU metrics. It is shown that improving both the STT system and the post-STT segmentations can lower the TER scores by up to 5.3% absolute and increase the BLEU scores by up to 2.7% absolute. © 2007 IEEE.

EARLY FLOWERING4 recruitment of EARLY FLOWERING3 in the nucleus sustains the Arabidopsis circadian clock

The plant circadian clock is proposed to be a network of several interconnected feedback loops, and loss of any component leads to changes in oscillator speed. We previously reported that Arabidopsis thaliana EARLY FLOWERING4 (ELF4) is required to sustain this oscillator and that the elf4 mutant is arrhythmic. This phenotype is shared with both elf3 and lux. Here, we show that overexpression of either ELF3 or LUX ARRHYTHMO (LUX) complements the elf4 mutant phenotype. Furthermore, ELF4 causes ELF3 to form foci in the nucleus. We used expression data to direct a mathematical position of ELF3 in the clock network. This revealed direct effects on the morning clock gene PRR9, and we determined association of ELF3 to a conserved region of the PRR9 promoter. A cis-element in this region was suggestive of ELF3 recruitment by the transcription factor LUX, consistent with both ELF3 and LUX acting genetically downstream of ELF4. Taken together, using integrated approaches, we identified ELF4/ELF3 together with LUX to be pivotal for sustenance of plant circadian rhythms. © 2012 American Society of Plant Biologists.

A Bayesian approach to tracking in single molecule fluorescence microscopy

Using fluorescence microscopy with single molecule sensitivity it is now possible to follow the movement of individual fluorophore tagged molecules such as proteins and lipids in the cell membrane with nanometer precision. These experiments are important as they allow many key biological processes on the cell membrane and in the cell, such as transcription, translation and DNA replication, to be studied at new levels of detail. Computerized microscopes generate sequences of images (in the order of tens to hundreds) of the molecules diffusing and one of the challenges is to track these molecules to obtain reliable statistics such as speed distributions, diffusion patterns, intracellular positioning, etc. The data set is challenging because the molecules are tagged with a single or small number of fluorophores, which makes it difficult to distinguish them from the background, the fluorophore bleaches irreversibly over time, the number of tagged molecules are unknown and there is occasional loss of signal from the tagged molecules. All these factors make accurate tracking over long trajectories difficult. Also the experiments are technically difficulty to conduct and thus there is a pressing need to develop better algorithms to extract the maximum information from the data. For this purpose we propose a Bayesian approach and apply our technique to synthetic and a real experimental data set.

Investigation of the effect of combustor cooling geometry on acoustic energy absorption

To control combustion instabilities occurring in LPP gas turbine combustors, several active and passive systems have been developed in recent years. The combustion chamber cooling geometry has the potential to influence instability feedback loops by absorbing acoustical energy inside the combustor. The design of the cooling liner and the geometry of the cooling plenum and the cooling air flow rate have a significant influence on the absorption characteristics of the system. This paper presents the results of a cold flow study which was carried out in the course of a comprehensive study on the influence of the cooling geometry on combustor thermoacoustics. Absorption characteristics of three different cooling liner geometries and non-perforated plates were determined over a frequency range from 50 Hz to 600 Hz for different cooling flow rates and different cooling plenum volumes. The experimental results compared well with results from a low order thermoacoustic network model. The acoustic energy absorption spectrum of a cooling liner with 90°-hole configuration was found to be strongly dependent on cooling flow rate and cooling plenum volume, whereas the absorption spectrum of cooling liners with 25°-holes were found to be strongly dependent on the cooling plenum volume, but less dependent on the cooling air flow rate. All cooling liner setups with perforations were capable of increased acoustic absorption over a broad band of frequencies compared to the case of non-perforated combustor walls. © 2010 by Johannes Schmidt.

Influencing factors of successful transitions towards product-service systems: A simulation approach

Product-Service Systems (PSS) are new business strategies moving and extending the product value towards its functional usage and related required services. From a theoretical point of view the PSS concept is known since a decade and many Authors reported reasonable possible success factors: higher profits over the entire life-cycle, diminished environmental burden, and localization of required services. Nevertheless the PSS promises remain quantitatively unproven relaying on a simple theory that involves a few constructs with some empirical grounding, but that is limited by weak conceptualization, few propositions, and/or rough underlying theoretical logic. A plausible interpretation to analyze the possible evolution of a PSS strategy could be considering it as a new business proposition competing on a traditional Product-Oriented (PO) market, assumed at its own equilibrium state at a given time. The analysis of the dynamics associated to a possible transition from a traditional PO to a PSS strategy allows investigating the main parameters and variables influencing an eventual successful adoption. This research is worthwhile because organizations undergoing fundamental PSS strategy are concerned about change and inertia key processes which, despite equilibrium theory and because of negative feedback loops, could undermine, economically, the return of their PSS proposition. In this paper Authors propose a qualitative System Dynamics (SD) approach by considering the PSS as a perturbation of an existing PO market featured by a set of known parameters. The proposed model incorporates several PSS factors able to influence the success of a PSS proposition under a set of given and justified assumptions, attempting to place this business strategy in a dynamic framework.

Discriminative language model adaptation for Mandarin broadcast speech transcription and translation

This paper investigates unsupervised test-time adaptation of language models (LM) using discriminative methods for a Mandarin broadcast speech transcription and translation task. A standard approach to adapt interpolated language models to is to optimize the component weights by minimizing the perplexity on supervision data. This is a widely made approximation for language modeling in automatic speech recognition (ASR) systems. For speech translation tasks, it is unclear whether a strong correlation still exists between perplexity and various forms of error cost functions in recognition and translation stages. The proposed minimum Bayes risk (MBR) based approach provides a flexible framework for unsupervised LM adaptation. It generalizes to a variety of forms of recognition and translation error metrics. LM adaptation is performed at the audio document level using either the character error rate (CER), or translation edit rate (TER) as the cost function. An efficient parameter estimation scheme using the extended Baum-Welch (EBW) algorithm is proposed. Experimental results on a state-of-the-art speech recognition and translation system are presented. The MBR adapted language models gave the best recognition and translation performance and reduced the TER score by up to 0.54% absolute. © 2007 IEEE.

Development of the 2003 CU-HTK conversational telephone speech transcription system

This paper describes the development of the 2003 CU-HTK large vocabulary speech recognition system for Conversational Telephone Speech (CTS). The system was designed based on a multi-pass, multi-branch structure where the output of all branches is combined using system combination. A number of advanced modelling techniques such as Speaker Adaptive Training, Heteroscedastic Linear Discriminant Analysis, Minimum Phone Error estimation and specially constructed Single Pronunciation dictionaries were employed. The effectiveness of each of these techniques and their potential contribution to the result of system combination was evaluated in the framework of a state-of-the-art LVCSR system with sophisticated adaptation. The final 2003 CU-HTK CTS system constructed from some of these models is described and its performance on the DARPA/NIST 2003 Rich Transcription (RT-03) evaluation test set is discussed.

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.

Mechanically tunable conjugated polymer distributed feedback lasers

Herein we report a low-threshold organic laser device based on semiconducting poly(9, 9′ -dioctylfluoren-2,7-diyl-alt-benzothiadiazole) (F8BT) encapsulated in a mechanically stretchable polydimethylsiloxane (PDMS) matrix. We take advantage of the natural flexibility of PDMS to alter the periodicity of the distributed feedback grating which in turn tunes the gain wavelength at which the resonant feedback is obtained. This way, we demonstrate that low-threshold lasing [6.1 μJ cm-2 (5.3 nJ)] is maintained over a large stretching range of 0%-7% which translates into a tuning range of about 20 nm. © 2010 American Institute of Physics.