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.

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.

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.

Extraordinarily thick-boned fish linked to the aridification of the Qaidam Basin (northern Tibetan Plateau)

Scattered with numerous salt lakes and approximate to 2,700-3,200 m above sea level, the giant Qaidam inland basin on the northern Tibetan Plateau has experienced continuing aridification since the beginning of the Late Cenozoic as a result of the India-Asia plate collision and associated uplift of the Tibetan Plateau. Previous evidence of aridification comes mainly from evaporite deposits and salinity-tolerant invertebrate fossils. Vertebrate fossils were rare until recent discoveries of abundant fish. Here, we report an unusual cyprinid fish, Hsianwenia wui, gen. et sp. nov., from Pliocene lake deposits of the Qaidam Basin, characterized by an extraordinarily thick skeleton that occupied almost the entire body. Such enormous skeletal thickening, apparently leaving little room for muscles, is unknown among extant fish. However, an almost identical condition occurs in the much smaller cyprinodontid Aphanius crassicaudus (Cyprinodonyiformes), collected from evaporites exposed along the northern margins of the Mediterranean Sea during the Messinian desiccation period. H. wui and A. crassicaudus both occur in similar deposits rich in carbonates (CaCO3) and sulfates (CaSO4), indicating that both were adapted to the extreme conditions resulting from the ariclification in the two areas. The overall skeletal thickening was most likely formed through deposition of the oversaturated calcium and was apparently a normal feature of the biology and growth of these fish.

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.

function definition language fdl and its implementation

A Function Definition Language (FDL) is presented. Though designed for describing specifications, FDL is also a general-purpose functional programming language. It uses context-free language as data type, supports pattern matching definition of functions, offers several function definition forms, and is executable. It is shown that FDL has strong expressiveness, is easy to use and describes algorithms concisely and naturally. An interpreter of FDL is introduced. Experiments and discussion are included.

yet another meta-language for programming language processing

The formal specification language LFC was designed to support formal specification acquisition. However, it is yet suited to be used as a meta-language for specifying programming language processing. This paper introduces LFC as a meta-language, and compares it with ASF+SDF, an algebraic specification formalism that can also be used to programming languages.

combining cfg and recursive functions to get a new language

We propose a new functional programming language(FPL) which differs in some aspects from most well known FPLs[l].We descrihc the prohlmm domain,the language,explain why we need it.

Gas-bearing fluid influx sub-system for gas hydrate geological system in Shenhu Area, Northern South China Sea

Based on the comprehensive interpretation and study of the Neogene fracture system and diapiric structure, it can be concluded that the diapiric structures, high-angle fractures and vertical fissure system are the main gas-bearing fluid influx sub-system for gas hydrate geological system in Shenhu Area, northern South China Sea. The Neogene fractures widely developed in the study area may be classed into two groups: NW (NNW)-trending and NE (NNE)-trending. The first group was active in the Late Miocene, while the second one was active since the Pliocene. The NE (NNE)-trending fractures were characterized by lower activity strength and larger scale, and cut through the sediment layers deposited since the Pliocene. Within the top sediment layers, the high-angle fracture and vertical fissure system was developed. The diapiric structures display various types such as a turtle-back-like arch, weak piercing, gas chimney, and fracture (or crack, fissure). On the seismic profile, some diapiric structures show the vertical chimney pathway whose top is narrow and the bottom is wide, where some ones extend horizontally into pocket or flower-shaped structures and formed the seismic reflection chaotic zones. Within the overlying sediment layers of the diapiric structure, the tree branch, flower-shaped high-angle fractures and vertical fissures were developed and became the pathway and migration system of the gas-bearing fluid influx. In the study area, the diapiric structures indicate a high temperature/over pressure system ever developed. Closely associated and abundant bright-spots show the methane-bearing fluid influx migrated vertically or horizontally through the diapiric structures, high-angle fractures and vertical fissures. In the place where the temperature and pressure conditions were favor for the formation of gas hydrate, the hydrate reservoir deposition sub-system was developed.