*** Call for Participation / Papers ***

International Workshop on Spoken Language Translation 2005

October 24-25, 2005, Pittsburgh, PA

http://www.is.cs.cmu.edu/iwslt2005/

The Consortium for Speech Translation Advanced Research (C-STAR), an international partnership of research laboratories engaged in automatic translation of spoken language, organizes The International Workshop on Spoken Language Translation (IWSLT) from Oct. 24th to 25th. Speech translation systems combine speech recognition (ASR) systems with machine translation (MT) system.

This year, IWSLT will focus on dealing with the integration of speech, and how to achieve more robust translations in face of recognition errors. An evaluation campaign will be held, using the multilingual BTEC corpus from the IWSLT-2004 evaluation, but extend it by using ASR output. In addition, multiple language trancks and integration modalities will be compared.

Two types of submissions are invited: 1) participants in the evaluation campaign of spoken language translation technologies, and 2) technical papers on related speech translation activities, including reports on work from other programs worldwide. Each participant in the evaluation campaign is requested to submit a paper describing the translation system.

Evaluation campaign

The evaluation of spoken language translation systems will be done using the Basic Travel Expression Corpus (BTEC). Participants choose at least on of the following conditions:

Translation directions:

• Chinese → English
• English → Chinese
• Japanese → English
• Korean → English
• Arabic → English

Data tracks:

• Supplied data: 20,000 sentence pairs for each translation direction

• Unrestricted data: supplied data, LDC data, WWW

• C-STAR track: Unrestricted data (including full BTEC corpus or proprietary data.)

Translation conditions:

• Manual transcription (plain sentences in BTEC)

• ASR output (1-best, N-best, lattices) of spoken BTEC sentences

Evaluation Methodology:

• Subjective Evaluation (most popular condition)

• Automatic Evaluation (BLEU, NIST, WER, METEOR, etc)

Technical Papers

To accomplish speech translation, we have to solve problems in MT for speech, especially for spontaneous speech and handle recognition errors in ASR output. Technical papers are invited on all matters related to speech translation. Potential topics for general speech translation session include, but are not limited to:

• Spontaneous speech translation

• Domain-limited and unlimited speech translation

• Tight coupling MT with ASR

• Word / Phrase Alignment

• MT Decoding Algorithms

• Multilingual ASR

• Multilingual Parsing

• Using comparable and non-parallel data

• Domain and Language Portability

• MT Evaluation Measures

Organizers:

Alex Waibel (CMU, USA / UKA, Germany; Chair)

Chiori Hori (CMU, USA, Co-Chair)

Stephan Vogel (CMU, USA, Program Chair)

Tanja Schultz (CMU, USA, Program Co-Chair)

Christian Boitet (CLIPS, France)

Gianni Lazzari (ITC-irst, Italy)

Youngjik Lee (ETRI, Korea)

Seiichi Yamamoto (ATR, Japan)

Chengqing Zong (CAS, China)

Celine Carraux (CMU, USA)

Contact:

InterACT, Language Technologies Institute, Carnegie Mellon University

TEL/FAX: +1-412-268-9177
E-mail: chiori@cs.cmu.edu

Sponsored and supported by the following organizations: