Victoria Arranz Corzana                    

I moved on January 2005 to:  ELDA/ELRA   (Evaluations and Language Resources Distribution Agency / European Language Resources Association) http:://www.elda.org arranz at elda <dot> org

Pen Picture
 

My first degree, from the University of Deusto, in Bilbao (Bizkaia), was in English Studies (Licenciatura en Filología Inglesa). I have an M.Sc. in Machine Translation from the Department of Language Engineering at UMIST (Manchester, UK), where I also received my Ph.D. in Computational Linguistics in 1998. My thesis was on Sublanguage-Based Semantic Clustering and Disambiguation from Corpora and I worked under the supervision of Prof. Jun-ichi Tsujii (UMIST and University of Tokyo). My work was within the framework of the Matsushita I (Methodologies for Development of Sublanguage Machine Translation) and Matsushita II (Exploiting Patterns for Hypertext Authoring and Navigation) projects.

After completing my Ph.D., I worked for a year at gilcUB (Grup d'Investigació en Lingüística Computacional - Universitat de Barcelona). For part of this time I was a researcher working on the SIMPLE (Semantic Information for Multifunctional Plurilingual Lexica) project. I also worked on the development of linguistic resources for Spanish.

During 1999 I also worked in CLIC (Centre de Lingüística i Computació - Universitat de Barcelona) on the development of linguistic resources for the medical sublanguage. This work took place within the SCRIPTUM project (Development of a Clinical Language Processor). Here I was once again in contact with the study of sublanguages and their terminology.

I am now at UPC (Universitat Politècnica de Catalunya), where I am a Research Associate and teach Computational Linguistics for doctoral students. I arrived in September 1999 to work on the BASURDE project (Spontaneus-Speech Dialogue System in Limited Domains). Through this project I had the opportunity to start working on dialogue systems, spontaneous speech and language understanding in a real system. Needless to say how much fun it can be trying to argue with a speech recognition system "who" insists on recognising Palencia instead of Valencia...oh well...

Then I moved onto the area of Speech-to-Speech Machine Translation (SST), which allowed me to undertake MT once again but applied to spoken language. We are working both on interlingua and statistics-based SST, exploring both approaches. This is taking place within several projects. One of these projects is FAME (Facilitating Agents for Multicultural Exchange), which aims at constructing an intelligent agent to facilitate communication among people at separate locations and from different cultures. A public demo took place at the Barcelona Forum of Cultures in July 2004 during the Language Technologies Day that was celebrated. A second demo also took place within Forum premises the following week in parallel with the Annual Meeting of the Association for Computation Linguistics. Both demos showed our latest technology in SST (with ASR, TTS and MT) for Catalan, English and Spanish.

The LC-STAR project (Lexica and Corpora for Speech to Speech Translation Components) is also within the field of SST. This project aims to develop language resources (lexica and corpora) for SST components, focusing on the current needs of existing SST technology researchers and developers. The results of LC-STAR will be made available via ELRA. We are also working on the ALIADO project (ALIADO: Tecnologías del Habla y el Lenguaje para un Asistente Personal), which undertakes the study and the development of spoken and written language technologies for the design of personal assistants with a mobile terminal in a multilingual environment. It considers two kinds of language-centred help that can be provied by the assistant: question answering and text or speech machine translation.

In relation to my work on SST, in 2002 I visited the LTI (Language Technologies Institute) and ISL (Interactive Systems Laboratories) at CMU (Carnegie Mellon University), Pittsburgh, USA, where I had the opportunity to work with interlingua-based machine translation experts such as Lori Levin, Donna Gates, Kay Peterson and Dorcas Wallace. ISL's Director Prof. Alex Waibel is co-founder of the International Consortium C-STAR (Consortium for Speech Translation Advanced Research) and also a member of the above-mentioned FAME project. Some of these colleagues also visited us at UPC in the late spring of 2003. In 2003 I also visited the Facultät für Informatik at the Universität Karlsruhe (UKA), Karsruhe, Germany, where there is the sister location of the Interactive Systems Labs at CMU.

I belong to the NLP group within the LSI (Software) department, which is also a member of the TALP research centre (Research Centre for Language and Speech Technology and Applications). TALP is also an affiliate member of the C-STAR and we are currently working on the development of SST technology for Catalan and Spanish.

Last but not least, I also lecture at the LSI department. I teach part of the Ph.D. course Fundamentals of Computational Linguistics. For those not familiar with the idea of lectures and doctoral research, in Spain you have to a ttend lectures before starting your actual research thesis. It kind of works like doing a Master before a Ph.D.
 

Publications

Teaching Activities

    Ph.D. Course Fundamentals of Computational Linguistics - FCL (2001-2002)

    Ph.D. Course Fundamentals of Computational Linguistics - FCL (2002-2003)

    Ph.D. Course Fundamentals of Computational Linguistics - FCL (2003-2004)

     Ph.D. Course Fundamentals of Computational Linguistics - FCL (2004-2005)

Research

(personal stuff)