1 Citation
This dataset includes 96kHz version of the CSTR VCTK Corpus including speech data uttered by 109 native speakers of English with various accents. The main dataset can be found at https://doi.org/10.7488/ds/1994 (containing a lower quality set of the same recordings) - for further details about this work please see the README.txt file and metadata at https://doi.org/10.7488/ds/1994.