The NCHLT speech corpus contains wide-band speech from approximately
200 speakers per language, in each of the eleven
official languages of South Africa. We describe the design and
development processes that were undertaken in order to develop
the corpus, and report on associated materials such as orthographic
transcriptions and pronunciation dictionaries that were
released as part of the corpus. In order to benchmark speechrecognition
performance on the corpus, we have also developed
both phone-recognition and word-recognition systems for all
eleven languages; we find that high accuracies can be achieved
for these speaker-independent but vocabulary-dependent recognition
tasks in all languages.