Information extraction and text generation of news reports for a Swedish-English bilingual spoken dialogue system
نویسندگان
چکیده
This paper describes an experimental dialog system designed to retrieve information and generate summaries of internet news reports related to user queries in Swedish and English. The extraction component is based on parsing and on matching the parsing output against stereotypic event templates. Bilingual text generation is accomplished by filling the templates after which grammar components generate the final text. The interfaces between the templates and the language-specific text generators are marked for prosodic information resulting in a text output where deaccentuation, accentuation, levels of focal accentuation, and phrasing are specified. These prosodic markers modify the default prosody rules of the text-to-speech system which then reads the text with subsequent improvement in intonation.
منابع مشابه
Spoken to Spoken vs. Spoken to Written: Corpus Approach to Exploring Interpreting and Subtitling
issue of Polibits includes a selection of papers related to the topic of processing of semantic information. Processing of semantic information involves usage of methods and technologies that help machines to understand the meaning of information. These methods automatically perform analysis, extraction, generation, interpretation, and annotation of information contained on the Web, corpus, nat...
متن کاملUsing text mining to identify crime patterns from Arabic crime news report corpus
Most text mining techniques have been proposed only for English text, and even here, most research has been conducted on specific texts related to special contexts within the English language, such as politics, medicine and crime. In contrast, although Arabic is a widely spoken language, few mining tools have been developed to process Arabic text, and some Arabic domains have not been studied a...
متن کاملBuilding English-Vietnamese Named Entity Corpus with Aligned Bilingual News Articles
Named entity recognition aims to classify words in a document into pre-defined target entity classes. It is now considered to be fundamental for many natural language processing tasks such as information retrieval, machine translation, information extraction and question answering. This paper presents a workflow to build an English-Vietnamese named entity corpus from an aligned bilingual corpus...
متن کاملSemi-Automatic Parallel Corpora Extraction from Comparable News Corpora
The parallel corpus is a necessary resource in many multi/cross lingual natural language processing applications that include Machine Translation and Cross Lingual Information Retreival. Preparation of large scale parallel corpus takes time and also demands the linguistics skill. In the present work, a technique has been developed that extracts parallel corpus between Manipuri, a morphologicall...
متن کاملOn-Line Learning of a Persian Spoken Dialogue System Using Real Training Data
The first spoken dialogue system developed for the Persian language is introduced. This is a ticket reservation system with Persian ASR and NLU modules. The focus of the paper is on learning the dialogue management module. In this work, real on-line training data are used during the learning process. For on-line learning, the effect of the variations of discount factor (g) on the learning speed...
متن کامل