Showing papers on "Head (linguistics) published in 2004"

Visual Prosody and Speech Intelligibility Head Movement Improves Auditory Speech Perception

Abstract: People naturally move their heads when they speak, and our study shows that this rhythmic head motion conveys linguistic information. Three-dimensional head and face motion and the acoustics of a talker producing Japanese sentences were recorded and analyzed. The head movement correlated strongly with the pitch (fundamental frequency) and amplitude of the talker's voice. In a perception study, Japanese subjects viewed realistic talking-head animations based on these movement recordings in a speech-in-noise task. The animations allowed the head motion to be manipulated without changing other characteristics of the visual or acoustic speech. Subjects correctly identified more syllables when natural head motion was present in the animation than when it was eliminated or distorted. These results suggest that nonverbal gestures such as head movements play a more direct role in the perception of speech than previously known.

Hand, head and face - negative constructions in sign languages

Abstract: This article presents a typology of negative constructions across a substantial number of sign languages from around the globe. After situating the topic within the wider context of linguistic typology, the main negation strategies found across sign languages are described. Nonmanual negation includes the use of head movements and facial expressions for negation and is of great importance in sign languages as well as particularly interesting from a typological point of view. As far as manual signs are concerned, independent negative particles represent the dominant strategy, but there are also instances of irregular negation in most sign languages. Irregular negatives may take the form of suppletion, cliticisation, affixing, or internal modification of a sign. The results of the study lead to interesting generalisations about similarities and differences between negatives in signed and spoken languages.

Enriching a French Treebank

Abstract: This paper presents the current status of the French treebank developed at Paris 7 (Abeille et al., 2003a). The corpus comprises 1 million words from the newspaper le Monde, fully annotated and disambiguated for parts of speech, inflectional morphology, compounds and lemmas, and syntactic constituents. It is representative of contemporary normalized written French, and covers a variety of authors and subjects (economy, literature, politics, etc.), with extracts from newspapers ranging from 1989 to 1993. It has been used by computational linguists to train and evaluate taggers, parsers and lemmatizers, as well as by psycholinguists to extract lexical and syntactic preferences (Pynte et al., 2001). It is now being enriched with functional information, and used for parsing evaluation. 1. The French treebank Similarly to the Penn TreeBank, we have annotated both parts of speech and constituents. Differently from the Penn Treebank, we have also annotated compounds, lemmas and inflectional morphology. Our annotation choices are meant to be linguistically motivated and compatible with various linguistic theories. We have chosen surface-based annotations, with no empty categories (Abeille and Clement, 2002; Abeille et al., 2003b; Abeille, 2003). With compounds amalgamated and not counting punctuation marks, the treebank comprises 870 000 tokens, using 37 000 different lemmas, making up about 32 000 independent sentences. The average number of words per sentence is 27 and the average number of phrases is 20 (some phrases are unary). It has been automatically tagged and hand-corrected by human annotators in a first phase, and automatically chunked and hand-corrected in a second phase (Clement, 2001; Toussenel, 2001; Abeille et al., 2003a). In the first phase, the task of the annotators was to validate the sentence boundaries, as well as the compounds (for missing compounds or possible compounds irrelevant in a given context), and to validate the morpho-syntactic tags, especially for notoriously difficult cases (for example as a preposition or as a determiner). In the second phase, the annotators’ task was to validate the constituant labels and boundaries, adding embedding where appropriate, as well as to signal remaining errors which could have been overlooked in the first phase. They used a specific Emacs-based annotation tool. The annotated and validated corpus is formatted in XML, using the XCES recommendations, and is available for research purposes. We distinguish 14 lexical categories, used for simple words as well as for compounds: A (adjective), Adv (Adverb), CC (coordinating conjunction), CL (weak clitic pronoun), CS (subordinating conjunction), D (determiner) ET (foreign word), I (interjection), NC (common noun), NP (proper name), P (preposition), PRO (strong pronoun), V (verb), PONCT (punctuation mark). We distinguish 12 phrasal categories: AP (adjectival phrase), AdP (adverbial phrase), COORD (coordinated phrase), NP (noun phrase), PP (preposition phrase), VN (verbal nucleus), VPinf (infinitival clause), VPpart (participial clause), SENT (independent clause), Sint (parenthetical), Srel (relative clause), Ssub (other subordinated clause) We chose to only annotate major phrases, with little internal structure (we have determiners and modifying adjectives at the same level in the noun phrase for example). For the sake of simplicity, we make a parsimonious use of unary phrases. For rigid sequences of categories, such as dates or titles, it is difficult to determine the head, and we have one global NP with no internal constituents. For coordinations, we have a COORD phrase, for the conjunction and the non initial conjuncts) usually included inside a major phrase (headed by the initial conjunct). We do not have discontinuous constituents, since these can usually be recovered at the functional level : in Combien voulez-vous de pommes (lit. how many do you want of apples ?) both and de pommes have the same Object function. Most of the difficult cases were with PP attachment, or scope of coordination, and human annotators had to spend the necessary time to fully understand the sentences. We got rid of spurious ambiguities (with the same interpretation) by a Attach high heuristics, for example in support verb constructions such as ecrire un livre sur les indiens (write a book about Indians) where the PP complement passes the linguistic tests both as a complement of the Verb and as a complement of the preceding Noun, with no semantic difference. 2. Enrichment of the treebank 2.1. Enriching the treebank with grammatical functions Similarly to what has been done for the German Negra or Tiger Treebanks (Brants et al., 2003), we have added some functional information to the French treebank. We chose to annotate surface grammatical functions only, and mark them as labels on the phrasal categories. For clitics, we mark the corresponding functions on the verbal nucleus. Functional information such as complement (or modifier) of Noun or complement of Adjective is already implicit in the constituent hierarchy (or in the constituent label for relative clauses). So we have concentrated on the

Economy, Innovation, and Prescriptivism: From Spec to Head and Head to Head

Abstract: This paper decribes cyclical changes in negative and wh-constructions as a change from Spec(ifier) to Head. It accounts for this change through an economy principle that says 'if possible, be a head'. The changes examined all show a tendency towards heads and head-checking but execute this in slightly different ways. In addition, innovations introduce new specifiers, and prescriptive rules retain them, counteracting the effects of economy. Changes from Head to Head and from Spec to Spec also occur. These proceed typically towards positions higher in the tree and can be explained via a 'merge over move' economy principle. The change involving heads I'll look at is the change of to from preposition to complementizer, and the changes involving specifiers involve French negatives and English relatives. Thus, certain instances of grammaticalization can be accounted for in structural terms.

An Approach to English Comparative Correlatives

Abstract: Recent syntactic theory has highlighted the importance of peripheral constructions such as the comparative correlative construction. This construction involves a pair of filler-gap constructions with unusual properties, where the first is a subordinate clause and the second a main clause. The construction has a number of related constructions. A version of HPSG, which assumes hierarchies of phrase types, can provide satisfactory analyses both for the comparative correlative construction and for the related constructions. The two clauses in the construction can be analysed as non standard head-filler phrases differing from standard head-filler phrases in certain respects. The construction as a whole can be analyzed as a non-standard head-adjunct phrase, in which the head and the phrase have different categories.

Are determiners heads

Abstract: The paper focuses on the relation between the determiner (D) and the common noun (N) in a noun phrase (NP). Four facts show that D depends on N: only N is relevant to whether NP can be used as an adjunct; possessive determiners are similar to clearly dependent possessives e.g. in Dutch and German; N decides whether or not D is obligatory; and in English only one D is possible per N. Three other facts show the converse, that N depends on D: in many languages D sometimes fuses with a preceding preposition (e.g. French de le = du; English for each = per); D decides whether or not N is obligatory; the ellipsis of N is a regular example of dependent ellipsis. Therefore D and N are mutually dependent, a relation which requires the structural flexibility offered by Word Grammar. This does not mean that NP has two heads, but rather that either D or N may be the head.

Support vector machines applied to the classification of semantic relations in nominalized noun phrases

Abstract: The discovery of semantic relations in text plays an important role in many NLP applications. This paper presents a method for the automatic classification of semantic relations in nominalized noun phrases. Nominalizations represent a subclass of NP constructions in which either the head or the modifier noun is derived from a verb while the other noun is an argument of this verb. Especially designed features are extracted automatically and used in a Support Vector Machine learning model. The paper presents preliminary results for the semantic classification of the most representative NP patterns using four distinct learning models.

Paul's Argument from Nature for the Veil in 1 Corinthians 11:13-15: A Testicle Instead of a Head Covering

Abstract: We are no longer able to decide with certainty which behavior Paul criticizes and which custom he means to introduce in 1 Cor 11:2–16. Traditionally, exegetes have conjectured that Paul was insisting that the pneumatic women leaders wear the veil according to Jewish custom. Yet, v. 15 maintains that women have their hair instead of a head-covering (peribolaivou), and thus militates against such an interpretation. In a very convoluted argument, which can no longer be unraveled completely, Paul adduces several points for “this custom” or hair fashion.2

System and method for creating a 3D figurine using 2D and 3D image capture

Abstract: A method of making a three dimensional head representative of a head of a three dimensional object includes the steps of creating a geometric file of the head of the three dimensional object, creating an image file of the head of the three dimensional object, milling a milled head based upon the geometric file, recasting the milled head into a sculptable head, sculpting the sculptable shape, painting features onto the sculpted shape based upon the image file.

Noun phrases without nouns

Abstract: In this paper, I investigate the theoretical status of noun phrases without nouns, i.e. noun phrases that do not contain a noun or pronoun, but only words that otherwise occur as modifiers of nouns. I investigate six possible analyses for such noun phrases: (1) that they are elliptical, (2) that the apparent modifiers are nouns, (3) that the apparent modifiers are heads, (4) that the determiner is the head, (5) that they are headless, (6) that all noun phrases are headless. Although the answers vary depending on the language investigated, I argue that the last hypothesis is generally the most plausible one.

Two structural positions for locative and directional PPs in Norwegian motion constructions

Abstract: This paper discusses two types of constructions in Norwegian where a combination of a verb of motion and a prepositional phrase are ambiguous between a reading of directed motion and a reading of located motion. Based on the differences in the syntactic behaviour of the two types of constructions with respect to a variety of tests (viz. VP constituency tests, adverbial placement, accent placement and the binding of anaphora), I argue that the two different readings have different argument structures and syntactic structures. On the directed motion reading, the PP appears low down in the verb phrase as complement to a functional head Path0, where it is interpreted as endpoint. Locative PPs, however, appear higher up in the structure as a verb phrase adjunct.

Between Shifts and Alternations: Ditransitive Constructions

Abstract: In this paper I present two classes of double object constructions in Modern Greek, i.e., the genitive, as well as the double accusative, ditransitive constructions. I show that these two classes differ from one another in that not both of them permit derivational processes such as the formation of adjectival passives. I also look at the case properties associated with the verbs which head Modern Greek genitive and double accusative ditransitive constructions. Finally, the analysis I propose for these constructions in Modern Greek are formalized using the Minimal Recursion Semantics (MRS) framework of Copestake et al. (2001) and Copestake et al. (2003).

Combination of facial movements on a 3D talking head

Abstract: Facial movements play an important role in interpreting spoken conversations and emotions. There are several types of movements, such as conversational signals, emotion displays, etc. We call these channels of facial movement. Realistic animation of these movements improves the realism, liveliness of the interaction between human and computers using embodied conversational agents. To date, no appropriate methods have been proposed for integrating all facial movements. We propose in this paper a scheme of combining facial movements on a 3D talking head. First, we concatenate the movements in the same channel to generate smooth transitions between adjacent movements. This combination only applies to individual muscles. The movements from all channels are then combined taking into account the resolution of possible conflicting muscles

East Nusantara: Genetic, Areal, and Typological Approaches

Abstract: Three approaches to language comparison are distinguished. An areal approach leads to the conclusion that East Nusantara and the Bird's Head constitute a linguistic area.

On the analysability of English exocentric compounds

Abstract: In the past two decades, numerous studies have been written on the successful application of metaphor, metonymy and blending in the analysis of idiomatic expressions which traditional linguistic literature treated as semantically unanalysable phenomena, that go against the theory of compositionality (on the nonanalysability of idioms see for example Allen 1986, Cruse 1991, Fraser 1970; on the analysability of idioms see for example Benczes 2002, Gibbs 1994, Lakoff 1987, Kovecses and Szabo 1996). A similar view was adopted for the so-called exocentric compound expressions (for the original definition of endoand exocentricity see Bloomfield 1933). Since the vast majority of English compounds is endocentric (Bloomfield 1933), linguistic literature has a tendency to mention exocentric combinations only peripherally (if they are mentioned at all), and views these constructions as semantically non-transparent (see for example Dirven and Verspoor 1998, Jespersen 1954, Katamba 1993, Levi 1978, Marchand 1960, Selkirk 1982, Spencer 1991). The present paper takes a close look at these much-ignored constructions and claims that the semantic relations that might hold between the modifier and head elements of such compounds are exactly the same

Syntactic and Semantic Mismatches in the Korean ko-Construction

Abstract: Traditional theories of coordination and subordination have focused on the syntactic structure in distinguishing the two constructions. Coordination is a structure in which two or more elements are joined in such a way that each of them can be the head of that structure. In contrast, subordination is a structure in which two elements are joined in such a way that one is dominated by the other. However, since Ross (1967), it has been observed that some coordinate sentences display typical asymmetrical characteristics (Goldsmith 1985, Postal 1993, Lakoff 1986). In explaining this, Culicover and Jackendoff (1997) proposed the possibility of syntactic and semantic mismatches. They argued that conditional and sentences, as in (2), are semantically asymmetrical while syntactically symmetrical.

Transducing Arabic Phrases into Head-Modifier (HM) Pairs for Arabic Information Retrieval

Abstract: In order to raise the precision of Information Retrieval systems, linguistically derived phrases may be used besides the traditional (single) keywords for both the document representation and the queries. Such phrases may either take the form of (monolithical) collocations or, as shown in this paper, of Head/Modifier pairs representing dependency structures in the text. In this paper we describe the rationale behind the Head/Modifier approach, which has been developed for English, and investigate its applicability to Arabic texts. In particular, we show how it has to be adapted in order to cope with the rather distinct syntactic particularities of the languages involved.