Show simple item record

Lwazi II Setswana TTS Corpus
Orthographic and phonemically aligned transcriptions.
Karen Calteaux
KCalteaux@csir.co.za
Meraka Institute, CSIR; North-West University
Creative Commons Attribution Non-Commercial No-Derivatives 3.0 South Africa license: https://creativecommons.org/licenses/by-nc-nd/3.0/
Setswana
Daniel van Niekerk; Georg Schlünz
Marelie Davel; Etienne Barnard; Charl van Heerden; Karen Calteaux; Febe de Wet; Carmen Moors; Bryan McAlister; Aditi Sharma Grover; Tebogo Reid.
https://hdl.handle.net/20.500.12185/451
977-713-789-932-1
Speech
Data
Monolingual speech corpora: Annotated
273 Mb (zipped)
1
2332 wav files with corresponding transcriptions.
Lwazi II
Books; Government Documents; Periodicals; Web
Phonetically balanced sentences were chosen to ensure adequate phone coverage, from the reference texts
Resource Catalogue
Resource Index
tsn
2018-02-05T20:22:02Z; 2018-03-05T17:54:11Z
2018-02-05T20:22:02Z; 2018-03-05T17:54:11Z
2015-11-20


Files in this item

Thumbnail

This item appears in the following Collection(s)

  • Resource Catalogue [350]
    A collection of language resources available for download from the RMA of SADiLaR. The collection mostly consists of resources developed with funding from the Department of Arts and Culture.
  • Resource Index [412]
    A collection of language resource metadata mostly collected during the NHN funded technology audit of 2009, as well as the SADiLaR technology audit of 2018. Not all resources in this collection are available for download.

Show simple item record