Sixth European Conference on Speech Communication and Technology

Budapest, Hungary
September 5-9, 1999

The Speechdat-Car Multilingual Speech Databases for In-Car Applications: Some First Validation Results

Henk van den Heuvel (1), Jerôme Boudy (2), Robrecht Comeyne (3), Stephan Euler (4), Asuncion Moreno (5), Gael Richard (2)

(1) SPEX, Nijmegen, Netherlands; (2) Matra Nortel Communications, Bois d’Arcy, France; (3) Lernout & Hauspie Speech Products, Ieper, Belgium; (4) Robert Bosch GmbH, R&D Division, Stuttgart, Germany; (5) UPC, Barcelona, Spain

The main objective of SpeechDat-Car is to develop a set of speech databases to support training and testing of multilingual speech recognition applications in the car environment. SpeechDat-Car started in April 1998 in the 4th EC framework under project code LE4-8334. The duration of the project is 30 months. Equivalent and similar resources for nine languages will be created: Danish, English, Finnish, Flemish/Dutch, French, German, Greek, Italian and Spanish. For each language 600 sessions will be recorded from at least 300 speakers. SpeechDat-Car commits itself to a strict validation protocol to ensure optimal quality and exchangeability of the databases. The first milestone in this respect is the validation of the recording platform and of a small subset of initial recordings. This paper briefly describes the database design and the recording platforms; next, it focuses on the objectives, the procedure, and some of the results of the early validation stage.

Full Paper (PDF)   Gnu-Zipped Postscript

Bibliographic reference.  Heuvel, Henk van den / Boudy, Jerôme / Comeyne, Robrecht / Euler, Stephan / Moreno, Asuncion / Richard, Gael (1999): "The speechdat-car multilingual speech databases for in-car applications: some first validation results", In EUROSPEECH'99, 2279-2282.