CIRCE Accented English Speech Stimuli Corpus

This dataset contains a corpus of audio accented speech stimuli in English, developed within the CIRCE project for use in perceptual experiments on accent perception and discrimination.The corpus includes 44 recordings from both male and female speakers who are native speakers of English (L1) as well as speakers who acquired English as a second language (L2). 4 accents of English as L1 and 11 accents of English as L2 are represented. All speakers read the same standardized text; individual recordings have an approximate duration of 23 seconds and are organized by speakers’ linguistic background (L1/L2).The dataset is designed to support matched-guise and verbal-guise experiments, as well as research and educational activities addressing accent perception, linguistic diversity, and accent discrimination. 

Important Notice on Access Conditions

The audio files contain identifiable voice data and are therefore subject to data protection and ethical regulations. For this reason, access to the recordings is restricted.

Access may be granted, upon request, exclusively for non-commercial scientific research purposes and in accordance with the informed consent provided by participants. Requests must specify the research objectives, institutional affiliation, and intended use of the materials, and are subject to approval by the designated data controllers.

Redistribution of the recordings and the creation of derivative works are not permitted. All access procedures are conducted in compliance with applicable data protection legislation, including EU Regulation 2016/679 (GDPR). Requests for access and further information should be addressed to the Contact Person indicated.