English accent sample recordings

Source: The Edinburgh International Accents of English Corpus (EdAcc).
Ramon Sanabria, Nikolay Bogoychev, Nina Markl, Andrea Carmantini, Ondrej Klejch and Peter Bell.
University of Edinburgh, Centre for Speech Technology Research.
Dataset: https://doi.org/10.7488/ds/7914
Official dataset distribution: https://huggingface.co/datasets/edinburghcstr/edacc
Paper: https://groups.inf.ed.ac.uk/edacc/edacc.pdf

Recordings and adapted MP3 samples: Creative Commons Attribution-ShareAlike 4.0 International.
License: https://creativecommons.org/licenses/by-sa/4.0/
Full license text is supplied as EDACC_CC_LICENSE.txt.

Changes: selected corpus sentence segments and Stella control-passage excerpts have been transcoded to MP3. The manifest records any excerpt boundaries. Spoken words have not been substituted. Display transcript casing/punctuation is normalized; exact corpus text is retained in the manifest.
Each sample's speaker ID, self-reported accent, dataset row and source link are recorded in english-accent-manifest.json. Accent descriptions are the speakers' own descriptions; they are not a claim that each recording defines a distinct canonical dialect.

No endorsement by the speakers, authors or University of Edinburgh is implied.