Abstract not found

Can concepts from the theory of neural computation contribute to formal theories of the mind? Recent research has explored the implications of one principle of neural computation, optimization, for the theory of grammar. Optimization over symbolic linguistic structures provides the core of a new grammatical architecture, optimality theory. The proposition that grammaticality equals optimality sheds light on a wide range of phenomena, from the gulf between production and comprehension in child language, to language learnability, to the fundamental questions of linguistic theory: What is it that the grammars of all languages share, and how may they differ? It is evident that the sciences of the brain and those of the mind are separated by many gulfs, not the least of which lies between the formal methods appropriate for continuous dynamical systems and those for discrete symbol structures. Yet recent research provides evidence that integration of

This paper presents a novel forrealization of optimality theory. Unlike pre- yions treatments of optimality in computational linguistics, starting with Ellison (1994), the new approach does not require any explicit marking and counting of constraint violations. It is based on the notion of "lenient composition", defined as the combination of ordinary composition and priority union. If an underlying form has outputs that can meet a given constraint, lenient composition enforces the constraint; ff none of the output candidates meets the constraint, lenient composition allows all of them. For the sake of greater efficiency, we may "eniently compose" the a. relation and all the constraints into a single finite-state transducer that maps each underlying form directly into its op- timal surface realizations, and vice versa.. Seen from this perspective, optimality theory is surprisingly similar to the two older strains of finite-state phonology: classical rewrite systems and two-level models. In particular, the ranking of optimality constraints corresponds to the ordering of rewrite rules.

From a logical perspective, categorial type systems can be situated within a landscape of substructural logics --- logics with a structure-sensitive consequence relation. Research on these logics has shown that the inhabitants of the substructural hierarchy can be systematically related by embedding translations on the basis of structural modalities. The modal operators offer controlled access to stronger logics from within weaker ones by licensing of structural operations. Linguistic material exhibits structure in dimensions not covered by the standard structural rules. The purpose of this paper is to generalize the modalisation and licensing strategy to two such dimensions: phrasal structure and headedness. Phrasal domain-sensitive type systems capture the notion of constituent structure; constituency relaxation can be licensed via an associativity modality. The opposition between heads and non-heads introduces dependency structure, an autonomous dimension of linguistic structure wh...

... document structure can be seen as an extension of Nunberg's `text-grammar'; it is also closely related to `logical' mark-up in languages like HTML and LATEX. We show that by using this intermediate representation, several subtasks in language generation and language understanding can be defined more cleanly.

Finite-state morphology has been successful in the description and computational implementa. tion of a wide variety of natural languages. However, the particular challenges of Arabic, and the limitations of some implementations of finite-state morphology, have led many researchers to believe that finite-state power was not sufficient to handle Arabic and other Semitic morphology. This paper illustrates how the morphotactics and the variation rules of Arabic have been described using only finitestate operations and how this approach has been implemented in a significant morphological analyzer/generator.

A fundamental debate in the machine learning of language has been the role of prior knowledge in the learning process. Purely nativist approaches, such as the Principles and Parameters model, build parameterized linguistic generalizations directly into the learning system. Purely empirical approaches use a general, domain-independent learning rule (Error Back-Propagation, Instance-based Generalization, Minimum Description Length) to learn linguistic generalizations directly from the data. In this paper we suggest that an alternative to the purely nativist or purely empiricist learning paradigms is to represent the prior knowledge of language as a set of abstract learning biases, which guide an empirical inductive learning algorithm. We test our idea by examining the machine learning of simple Sound Pattern of English ( S P E)-style phonological rules. We represent phonological rules as finite-state transducers that accept underlying forms as input and generate surface forms as output. We show that OSTIA, a general-purpose transducer induction algorithm, was incapable of learning simple phonological rules like flapping. We then augmented OSTIA with three kinds of learning biases that are specific to natural language phonology, and that are assumed explicitly or implicitly by every theory of phonology: faithfulness (underlying segments

Pronunciation by analogy (PbA) is a data-driven method for relating letters to sound, with potential application to next-generation text-to-speech systems. This paper extends previous work on PbA in several directions. First, we have included `full' pattern matching between input letter string and dictionary entries, as well as including lexical stress in letter-to-phoneme conversion. Second, we have extended the method to phonemeto -letter conversion. Third, and most important, we have experimented with multiple, different strategies for scoring the candidate pronunciations. Individual scores for each strategy are obtained on the basis of rank and either multiplied or summed to produce a final, overall score. Five strategies have been studied and results obtained from all 31 possible combinations. The two combination methods perform comparably, with the product rule only very marginally superior to the sum rule. Nonparametric statistical analysis reveals that performance improves as more strategies are included in the combination: this trend is very highly significant ( p 0 0005). Accordingly for letter-to-phoneme conversion, best results are obtained when all five strategies are combined: word accuracy is raised to 65.5% relative to 61.7% for our best previous result and 63.0% for the best-performing single strategy. These improvements are very highly significant ( p 0 and p 0 00011 respectively). Similar results were found for phoneme-to-letter and letter-to-stress conversion, although the former was an easier problem for PbA than letter-to-phoneme conversion and the latter was harder. The main sources of error for the multi-strategy approach are very similar to those for the best single strategy, and mostly involve vowel letters and phonemes. 1

This paper discusses a perceptron model of the learning and assignment of linguistic stress, using data from nineteen human languages. First, we point out some interesting parallels between aspects of the model and the constructs and predictions of metrical phonology, the linguistic theory of stress. Second, we develop a novel analysis of linguistic stress in terms of ease of perceptron-learnability. These two sets of results suggest that simple statistical learning techniques have the potential to complement, and provide computational validation for, abstract theoretical investigations of language. We then examine why such methodologies should be of interest for linguistic theorizing. Our analysis began at a high level by observing inherent characteristics of various stress systems, much as theoretical linguistics does. However, our explanations changed substantially whenwe included a detailed account of the model's processing mechanisms. Our higher-level, theoretical accou...

This paper describes the latest version of the SOLE concept-tospeech system, which uses linguistic information provided by a natural language generation system to improve the prosody of synthetic speech. We discuss the types of linguistic information that prove most useful and the implications for text-to-speech systems. 1. INTRODUCTION The purpose of the SOLE project is to make use of automaticallygenerated, high-level linguistic information to improve the quality of the intonation of synthetic speech. After choosing an initial set of linguistic constructs thought to have some influence on prosody, we developed an SGML-based mark-up language to serve as a general interface between NLG and speech synthesis systems, and trained our synthesis system to recognise correlations between the mark-up and intonational contours so that it can make use of this mark-up when synthesising. As a result, many of the errors that the synthesiser makes with regard to knowing when to accent or deaccent ...