In the context of spoken dialogue systems, we investigated a bottom-up robust parsing for LTAG (Lexicalized Tree Adjoining Grammars) that interleaves a syntactic and a semantic structure. When the regular syntactic composition rules fail, the syntactic islands and the corresponding partial semantic structures are combined thanks to additional local rules. We supply some descriptive limits of th...

The paper introduces our strategy for adapting a rule based parser of written language to transcribed speech. Special attention has been paid to disfluencies (repairs, repetitions and false starts). A Constraint Grammar based parser was used for shallow syntactic analysis of spoken Estonian. The modification of grammar and additional methods improved the recall from 97.5% to 97.7% and precision...

Abstract The dependency-based annotation format employed within the Spoken Dutch Corpus (CGN) project (van der Wouden et al., 2002) has been designed in such a way as to enable a transparent mapping to the derivational structures of current ‘lexicalized’ grammar formalisms. Through such translations, the CGN tree bank can be used to train and evaluate computational grammars within these framewo...

Spoken language systems, ranging from interactive voice response (IVR) to mixed-initiative conversational systems, make use of a wide range of recognition grammars and vocabularies. The recognition grammars are either static (created at design time) or dynamic (dependent on database lookup at run time). This paper examines the compilation of recognition grammars with an emphasis on the dynamic ...

We describe the methods and hardware that we are using to produce a real-time demonstration of an integrated Spoken Language System. We describe algorithms that greatly reduce the computation needed to compute the N-Best sentence hypotheses. To avoid grammar coverage problems we use a fully-connected first-order statistical class grammar. The speech-search algorithm is implemented on a board wi...

In this paper we propose a semantically structured language (SSLM) model that significantly reduces the authoring load required over the traditional manually derived grammar when developing a spoken language system. At the same time, the SSLM results in an understanding error rate which is roughly half as large as that of the manually authored grammar. The proposed model combines the advantages...

OBJECTIVES To assess the development of grammar comprehension in spoken language in prelingually deaf children following cochlear implantation and compare their grammatical abilities with those of their hearing peers. DESIGN A prospective study of 82 consecutive prelingually deaf children up to 5 years following implantation. The children were less than 7 years old at the time of implantation...

This paper reports on recent developments for the creation and analysis of very large databases of emotional and attitudinallymarked speech for the support of research into concatenative methods for producing synthesised speech which is capable of expressing the range of prosody and phonation styles to emulate human spoken interactions. It addresses the problems of ensuring high spontaneity in ...

This paper presents PARSEC-a system for generating connectionist parsing networks from example parses. PARSEC is not based on formal grammar systems and is geared toward spoken language tasks. PARSEC networks exhibit three strengths important for application to speech processing: 1) they learn to parse, and generalize well compared to handcoded grammars; 2) they tolerate several types of noise;...

This paper is a brief summary of some of our recent research on the processing of nominal classifiers in two Sinitic languages spoken in Taiwan, Mandarin and Taiwanese (Southern Min Chinese). Our research has used data from classifier systems to address two interface issues, namely the relation between grammar and extra-grammatical cognition and between grammar and the lexicon. We have been res...