EUR-ACCOR

General

Authors: European consortium

Funded by: European ESPRIT grant number:

When created: 1990-1993

Availability: English speakers available on CDROM through ELRA for 25 ECU to members or 75 ECU to non members

Purpose: Cross-language investigation of acoustic-articulatory correlations in coarticulatory processes

Description: Overview

Instrumentation

Components:

• Microphone 20kHz sample rate (Sennheiser MKH 40 P48)
• Laryngograph 10kHz sample rate
• Rothenberg mask 500Hz sample rate
• EPG 200Hz sample rate

Corpus

Texts: A common corpus was used for all languages (with exceptions when sequences were not phonotactically permissable). The main speech materials were as follows:

1. Nonsense items:
   • Vowels /i,a,u/ in isolation
   • VCV sequences, where C= /p,tb,t,d,k,s,z,n,l,S,T/ and the sequences /kl,st/; V = /i,a,u/.
2. Real words:
   • These match the nonsense sequences above as closely as possible. eg. Nonsense item /iti/ is matched by the english word "meaty".
3. Sentences:
   • A set of 14 short sentences designed to illustrate the main connected speech processes in the language (eg. assimilations, weak forms ..). In some languages, items from the real word corpus appear in the sentences.
   Orthography

Subjects: 5-10 speakers per language each recorded 10 repetitions of the corpus

Conditions:

5 repetitions recorded with EPG, Laryngograph and audio.

5 repetitions recorded with EPG, Laryngograph, audio and nasal/oral airflow ( Rothenberg mask).

All recordings made in the same sound damped studio at the university of Reading. All data were recorded direct to computer with the exception of audio which was recorded onto DAT and later transferred.

Languages: Catalan,English,French,German,Irish Gaelic,Italian,Swedish

Examples: sound files and EDIT screen dump

Access

Platforms: Database access software currently runs on PC

Media: Planned release on CDROM

Format:

File format is currently under review and will be stored with NIST headers

Size:12.5Gbytes

Software:

• SPLIT - split files in old compressed format
• EDIT - simultaneous display of all track files in old format.
• New management software planned for new file format

Software Documentation: EDIT

Software Source/Executables:EDIT

Further information

Contact: A. Wrench

Address:

Dept. Speech and language Sciences, Queen Margaret College, Clerwood Terrace, Edinburgh. EH12.

Telephone:

+44 131 317 3692

Fax:

+44 131 317 3689

Email:

a.wrench@sls.qmced.ac.uk

WWW:

http://www.qmced.ac.uk