Search

In the later years the amount of freely available multilingual corpora has grown in an exponential way. Unfortunately the way these corpora are made available is very diverse, ranging from simple text files or specific XML schemas to supposedly standard formats like the XML Corpus Encoding Initiative, the Text Encoding Initiative, or even the Translation Memory Exchange formats. In this document we defend the usage of Translation Memory Exchange documents, but we enrich its structure in order to support the annotation of the documents with different information like lemmas, multi-words or entities. To support the adoption of the proposed formats, we present a set of tools to manipulate the different formats in an agile way.

The eXtensible Mark-up Language (XML) is probably one of the most popular markup languages available today. It is very typical to find all kind of services or programs representing data in this format. This situation is even more common in web development environments or Service Oriented Architectures (SOA), where data flows from one service to another, being consumed and produced by an heterogeneous set of applications, which sole requirement is to understand XML. This workflow of data represented in XML implies some tasks that applications have to perform if they are required to consume or produce information: the task of parsing an XML document, giving specific semantics to the information parsed, and the task of producing an XML document. Our main goal is to create object definitions that can analyze an XML document and automatically create an object definition that can be used abstractly by the application. These objects are able to parse the XML document and gather all the data required to mimic all the information present in the document. This paper introduces xml2pm, a simple tool that can inspect the structure of an XML document and create an object definition (a Perl module) that stores the same information present in the orinial document, but as a runtime object. We also introduce a simple case of how this approach allows the creation of applications based on Web Services in an elegant and simple way.

In our days, the notion, the importance and the significance of parallel corpora is so big that needs no special introduction. Unfortunately, public available parallel corpora is somewhat limited in range. There are big corpora about politics or legislation, about medicine and other specific areas, but we miss corpora for other different areas. Currently there is a huge investment on using the Web as a corpus. This article uncovers GWB, a tool that aims automatic construction of parallel corpora from the web. We defend that it is possible to build high quality terminological corpora in an automatic fashion, just by specifying a sensible Internet domain and using an appropriate set of seed keywords. GWB is a web-spider that works in conjunction with a set of other Open-Source tools, defining a pipeline that includes the documents retrieval from the web, alignment at sentence level and its quality analysis, bilingual dictionaries and terminology extraction and construction of off-line dictionaries.

Parallel corpora are rich sources of translation resources. This document presents a methodology for the extraction of bilingual nominals (terminology candidates) from parallel corpora, using translation patterns. The patterns proposed in this work specify the order changes that occur during translation and that are intrinsic to the involved languages syntaxes. These patterns are described in a domain specific language named PDL (Pattern Description Language), and are extremely efficient for the detection of nominal phrases., Los corpora paralelos son fuentes ricas en recursos de traducción. Este documento presenta una metodología para la extracción de sintagmas nominales bilingües (candidatos terminológicos) a partir de corpora paralelos, utilizando reglas de traducción. Los modelos propuestos en este trabajo especifican las alteraciones en el orden de las palabras que se producen durante la traducción y que son intrínsecos a la sintaxis de las lenguas implicadas. Estas reglas se describen en un lenguaje de dominio específico llamado PDL (Pattern Description Language) y son sumamente eficientes para la detección de sintagmas nominales., Alberto Simoes has a scholarship from Fundacao para a Computacao Cientifica Nacional and the work reported here has been partially funded by Fundacao para a Ciencia e Tecnologia through project POSI/PLP/43931/2001, co-financed by POSI, and by POSC project POSC/339/1.3/C/NAC.

The creation of a CGI or a WebService as an interface for a command line tool is not as unusual as it may seem. It is extremely usual and useful. There are applications developed as command line tools that can be useful for different purposes, and different kind of users. Some of these users might not be able to run these tools directly. For instance, it is not easy to install a bunch of Perl modules to have a small tool working. For these situations, it is easier to make the tool available in the Web or as a WebService. The problem with making the tool available in these fashions, is that programmers tend to rewrite the tools to incorporate the CGI or XML specific layers. We defend that these CGI or WebService interfaces should use the already available command line tool, without any change. This interface should be able to read a simple textual specification of how the command line tool works, and buid the CGI or XML specific layers automatically. The CGI::Auto module aims this purpose: to encapsulate command line tools in a CGI layer based on a textual specification, transforming the command line tool in a web application.

In this document we describe the use of Text : Rewrite Rules in several tasks related to Speech.

This paper describes NATools, a toolkit to process, analyze and extract translation resources from Parallel Corpora. It includes tools like a sentence-aligner, a probabilistic translation dictionaries extractor, word-aligner, a corpus server, a set of tools to query corpora and dictionaries, as well as a set of tools to extract bilingual resources., Alberto Simoes has a scholarship from Fundacao para a Computacao Cientifica Nacional and the work reported here has been partially funded by Fundacao para a Ciencia e Tecnologia through project POSI/PLP/43931/2001, co-financed by POSI, and by POSC project POSC/339/1.3//NAC.

Parallel corpora are important resources for most Natural Language processing tasks. From the common applications, like machine translation, to the usually mono-lingual tasks as paraphrase detection and word sense disambiguation, most researchers are using massive parallel corpora. Thus, the availability of an efficient way to manage them is very important. This paper presents a Client-Server architecture to query efficiently parallel corpora and probabilistic translation dictionaries., Los corpora paralelos son importantes para la mayoría de las tareas de procesamiento de Lenguaje Natural. Gran parte de los investigadores utilizan los corpora paralelos empleando-los en aplicaciones comunes, como la traducción máquina, y en tareas mono-linguísticas, como sean la detección de paráfrasis y la resolución de problemas de ambiguedad semántica. Este trabajo ha sido orientado por estas cuestiones y propone una arquitectura cliente-servidor de interrogación eficiente de corpora paralelos y diccionarios de traducción probabilística.

In this document we present an open source Portuguese text to speech. Our first goal is to provide a flexible way to extend it, using a generic way to convert Portuguese words on SAMPA phonemes, and consult dictionaries only on exceptions examples. The Text-to-Speech is compound of five layers, each one based on simple rules in a way to be easily tuned. In order to do that, we wrote a generic text rewriting system that is presented in the section two. The result of this work is a tool that can be used as an independent Text-to-Speech system or as a Natural Language Processing library for various tasks. We present some examples how them can be used in the Applications section.

In this paper we present a Perl module, called XML::DT, that can be used to translate and transform XML documents. A programmer always looks for the simplest tool to do a certain task, development and maintenance will be easier. That is the main idea behind the module we are presenting: a simple tool that will enable the user to speed up his work and that will help him to maintain it. XML::DT includes some down translation features that are common to other SGML/XML processors available on the market like Omnimark, or Balise, and some other features to deal with input and output of Unicode character sets. The idea was to adopt familiar concepts together with a familiar syntax to SGML/XML programmers but shaped to the usual Perl notation., GCA

Dictionary and Thesaurus are valuable resources for Natural Language Processing but do not exist as freely available as expected, especially for languages other than English and, when they exist, they are just available for querying online. Our main goal with T2O - Thesaurus to Ontology framework - is to create a multilingual ontology: freely available online and to download; with a computer readable format; with a good API; with a structure as rich as possible; reusing all the structured information we can get, (undefined)

Music sources are most commonly shared in music scores scanned or printed on paper sheets. These artifacts are rich in information, but since they are images it is hard to re-use and share their content in todays’ digital world. There are modern languages that can be used to transcribe music sheets, this is still a time consuming task, because of the complexity involved in the process and the typical huge size of the original documents. WIKI::SCORE is a collaborative environment where several people work together to transcribe music sheets to a shared medium, using the notation. This eases the process of transcribing huge documents, and stores the document in a well known notation, that can be used later on to publish the whole content in several formats, such as a PDF document, images or audio files for example., (undefined)