Word boundary modelling and full covariance gaussians for Arabic Speech-to-Text systems

This paper describes recent improvements to the Cambridge Arabic Large Vocabulary Continuous Speech Recognition (LVCSR) Speech-to-Text (STT) system. It is shown that wordboundary context markers provide a powerful method to enhance graphemic systems by implicit phonetic information, improving the modelling capability of graphemic systems. In addition, a robust technique for full covariance Gaussian modelling in the Minimum Phone Error (MPE) training framework is introduced. This reduces the full covariance training to a diagonal covariance training problem, thereby solving related robustness problems. The full system results show that the combined use of these and other techniques within a multi-branch combination framework reduces the Word Error Rate (WER) of the complete system by up to 5.9% relative. Copyright © 2011 ISCA.

IPLR: An online resource for Greek word-level and sublexical information

We present a new online psycholinguistic resource for Greek based on analyses of written corpora combined with text processing technologies developed at the Institute for Language & Speech Processing (ILSP), Greece. The "ILSP PsychoLinguistic Resource" (IPLR) is a freely accessible service via a dedicated web page, at http://speech.ilsp.gr/iplr. IPLR provides analyses of user-submitted letter strings (words and nonwords) as well as frequency tables for important units and conditions such as syllables, bigrams, and neighbors, calculated over two word lists based on printed text corpora and their phonetic transcription. Online tools allow retrieval of words matching user-specified orthographic or phonetic patterns. All results and processing code (in the Python programming language) are freely available for noncommercial educational or research use. © 2010 Springer Science+Business Media B.V.

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.

Using sub-word-level information for confidence estimation with conditional random field models

The task of word-level confidence estimation (CE) for automatic speech recognition (ASR) systems stands to benefit from the combination of suitably defined input features from multiple information sources. However, the information sources of interest may not necessarily operate at the same level of granularity as the underlying ASR system. The research described here builds on previous work on confidence estimation for ASR systems using features extracted from word-level recognition lattices, by incorporating information at the sub-word level. Furthermore, the use of Conditional Random Fields (CRFs) with hidden states is investigated as a technique to combine information for word-level CE. Performance improvements are shown using the sub-word-level information in linear-chain CRFs with appropriately engineered feature functions, as well as when applying the hidden-state CRF model at the word level.

Award Winner in the Young Investigator Category, 2014 Society for Biomaterials Annual Meeting and Exposition, Denver, Colorado, April 16-19, 2014: Periodically perforated core-shell collagen biomaterials balance cell infiltration, bioactivity, and mechanical properties.

Orthopedic tissue engineering requires biomaterials with robust mechanics as well as adequate porosity and permeability to support cell motility, proliferation, and new extracellular matrix (ECM) synthesis. While collagen-glycosaminoglycan (CG) scaffolds have been developed for a range of tissue engineering applications, they exhibit poor mechanical properties. Building on previous work in our lab that described composite CG biomaterials containing a porous scaffold core and nonporous CG membrane shell inspired by mechanically efficient core-shell composites in nature, this study explores an approach to improve cellular infiltration and metabolic health within these core-shell composites. We use indentation analyses to demonstrate that CG membranes, while less permeable than porous CG scaffolds, show similar permeability to dense materials such as small intestine submucosa (SIS). We also describe a simple method to fabricate CG membranes with organized arrays of microscale perforations. We demonstrate that perforated membranes support improved tenocyte migration into CG scaffolds, and that migration is enhanced by platelet-derived growth factor BB-mediated chemotaxis. CG core-shell composites fabricated with perforated membranes display scaffold-membrane integration with significantly improved tensile properties compared to scaffolds without membrane shells. Finally, we show that perforated membrane-scaffold composites support sustained tenocyte metabolic activity as well as improved cell infiltration and reduced expression of hypoxia-inducible factor 1α compared to composites with nonperforated membranes. These results will guide the design of improved biomaterials for tendon repair that are mechanically competent while also supporting infiltration of exogenous cells and other extrinsic mediators of wound healing.

The Category-Theoretic Solution of Recursive Domain Equations

Recursive specifications of domains plays a crucial role in denotational semantics as developed by Scott and Strachey and their followers. The purpose of the present paper is to set up a categorical framework in which the known techniques for solving these equations find a natural place. The idea is to follow the well-known analogy between partial orders and categories, generalizing from least fixed-points of continuous functions over cpos to initial ones of continuous functors over $\omega $-categories. To apply these general ideas we introduce Wand's ${\bf O}$-categories where the morphism-sets have a partial order structure and which include almost all the categories occurring in semantics. The idea is to find solutions in a derived category of embeddings and we give order-theoretic conditions which are easy to verify and which imply the needed categorical ones. The main tool is a very general form of the limit-colimit coincidence remarked by Scott. In the concluding section we outline how compatibility considerations are to be included in the framework. A future paper will show how Scott's universal domain method can be included too.

smoothing document language model with local word graph

ACM SIGIR; ACM SIGWEB

左侧梭状回内文字的字形加工：中文正常阅读者，中文文盲、非中文母语被试的对比研究

Reading is an important human-specific skill obtained through extensive learning experience and is reliance on the ability to rapidly recognize single words. According to the behavioral studies, the most important stage of reading is the representation of “visual word form”, which is independent on surface visual features of the reading materials. The prelexical visual word form representation is characterized by the abstractive and highly effective and precise processing. Neuroimaging and neuropsychological studies have investigated the neural basis underlying the visual word form processing. On the basis of summary of the existing literature, the current thesis aimed to address three fundamental questions involving neural basis of word recognition. First, is there a dedicated neural network that is specialized for word recognition? Second, is the orthographic information represented in the putative word/character selective region (VWFA)? Third, what is the role of reading experience in the genesis of the VWFA, is experience a main driver to shape VWFA instead of evolutionary selectivity? Nineteen Chinese literate volunteers, 5 Chinese illiterates and 4 native English speakers participated in this study, and performed perceptual tasks during fMRI scanning. To address the first question, we compared the differential responses to three categories of visual objects, i.e., faces, line drawings of objects and Chinese characters, and defined the region of interesting (ROI) for the next experiment. To address the second question, Chinese character orthography was manipulated to reveal possible differential responses to real characters, false characters, radical combinations, and stroke combinations in the regions defined by the first experiment. To examine the role of reading experience in genesis of specialization for character, the responses for unfamiliar Chinese characters in Chinese illiterates and native English speakers were compared with that in the Chinese literates, and tracked the change in cortical activation after a short-term reading training in the illiterates. Data were analyzed in two dimensions. Both BOLD signal amplitude and spatial distribution pattern among multi-voxels were used to systematically investigate the responsiveness of the left fusiform gyrus to Chinese characters. Our results provide strong and clear evidence for the existence of functionally specialized regions in the human ventral occipital-temporal cortex. In the skilled readers a region specialized for written words could be consistently found in the lateral part of the left fusiform gyrus, line drawings in the median part and faces in the middle. Our results further show that spatial distribution analysis, a method that was not commonly used in neuroimaging of reading, appears to be a more effective measurement for category specialization for visual objects processing. Although we failed to provide evidence that VWFA processes orthographic information in terms of signal intensitiy, we do show that response pattern of real characters and radical collections in this area is different from that of false characters and random stroke combinations. Our last set of experiments suggests that the selective bias to reading material is clearly experience dependent. The response to unknown characters in both English speakers/readers and Chinese illiterates is fundamentally different from that of the skilled Chinese readers. The response pattern for unknown characters is more similar to that for line drawings rather as a weak version of character in skilled Chinese readers. Short-term training is not sufficient to produce VWFA bias even when tested with learned characters, rather the learned characters generated a overall upward shift of the activation of the left fusiform region. Formation of a dedicated region specialized for visual word/character might depend on long-term extensive reading experience, or there might be a critical period for reading acquisition.

品牌属性权重判断的反应时研究

In current days, many companies have carried out their branding strategies, because strong brand usually provides confidence and reduce risks to its consumers. No matter what a brand is based on tangible products or services, it will possess the common attributes of this category, and it also has its unique attributes. Brand attribute is defined as descriptive features, which are intrinsic characteristics, values or benefits endowed by users of the product or service (Keller, 1993; Romaniuk, 2003). The researches on models of brand multi-attributes are one of the most studied areas of consumer psychology (Werbel, 1978), and attribute weight is one of its key pursuits. Marketing practitioners also paid much attention to evaluations of attributes. Because those evaluations are relevant to the competitiveness and the strategies of promotion and new product development of the company (Green & Krieger, 1995). Then, how brand attributes correlate with weight judgments? And what features the attribute judgment reaction? Especially, what will feature the attribute weight judgment process of consumer who is facing the homogeneity of brands? Enlightened by the lexical hypothesis of researches on personality traits of psychology, this study choose search engine brands as the subject and adopt reaction time, which has been introduced into multi-attributes decision making by many researchers. Researches on independence of affect and cognition and on primacy of affect have cued us that we can categorize brand attributes into informative and affective ones. Meanwhile, Park has gone further to differentiate representative and experiential with functional attributes. This classification reflects the trend of emotion-branding and brand-consumer relationship. Three parts compose the research: the survey to collect attribute words, experiment one on affective primacy and experiment two on correlation between weight judgment and reaction. The results are as follow: In experiment one, we found: (1) affect words are not rated significantly from cognitive attributes, but affect words are responded faster than cognitive ones; (2) subjects comprehend and respond in different ways to functional attribute words and to representative and experiential words. In experiment two, we fund: (1) a significant negative correlation between attributes weight judgment and reaction time; (2) affective attributes will cause faster reaction than cognitive ones; (3) the reaction time difference between functional and representative or experiential attribute is significant, but there is no different between representative and experiential. In sum, we conclude that: (1): In word comprehension and weight judgment, we observed the affective primacy, even when the affect stimulus is presented as meaningful words. (2): The negative correlation between weight judgment and reaction time suggest us that the more important of attribute, the quicker of the reaction. (3): The difference on reaction time of functional, representative and experiential reflects the trend of emotional branding.

CARPS: A Program which Solves Calculus Word Problems

A program was written to solve calculus word problems. The program, CARPS (CALculus Rate Problem Solver), is restricted to rate problems. The overall plan of the program is similar to Bobrow's STUDENT, the primary difference being the introduction of "structures" as the internal model in CARPS. Structures are stored internally as trees. Each structure is designed to hold the information gathered about one object. A description of CARPS is given by working through two problems, one in great detail. Also included is a critical analysis of STUDENT.