EUROM_1 European Languages Speech Database

One of the major tasks of the working group "Enabling Technology and Research" in the ESPRIT-project 2589 (SAM) was the production of standard European speech databases. Within the SAM project, standard recording protocols and several tools for recording and annotion of speech databases were developed. The whole EUROM.1 database covers eight European languages (Danish, Dutch, British English, French, German, Italian, Norwegian and Swedish). The Italian database is available separately, the other languages are described below.

Description

Overview

Languages:

Danish, Dutch, English, French, German, Norwegian, Swedish

Speakers:

60 speakers per language

Protocol:

20KHz 16bit sampling, anechoic room

Media:

5 CDROMs per language

Content: (for each language)

Many Talker Corpus (30 women, 30 men), 100 numbers, 3 passages, 5 sentences, (speech signal)

Few Talker Corpus (5 women and 5 men), 100 numbers x 5, 15 passages, 25 sentences, C(C)VC(V) x 5, (speech + laryngographic signals)

Very Few Talker Corpus (1 woman and 1 man), C(C)VC(V) material embedded in 5 context, phrases., Context words x 5, (speech + laryngographic signals)

Reference

D. Chan, A. Fourcin, D. Gibbon, B. Granstrom, M. Huckvale, G. Kokkinakis, K. Kvale, L. Lamel, B. Lindberg, A. Moreno, J. Mouropoulos, F. Senia, I. Trancoso, C. Veld & J. Zeiliger, "EUROM- A Spoken Language Resource for the EU", in Eurospeech'95. Proceedings of the 4th European Conference on Speech Communication and Speech Technology. Madrid, Spain, 18-21 September, 1995. Vol 1, pp. 867-870

Documentation

Overview

• Introduction to EUROM_1
• General technical information

Languages

Availability

EUROM_1 CD-ROMs are available for order from the European Language Resources Association.