SADiLaR Language Resource Repository
Not a member yet
536 research outputs found
Sort by
Sesotho function word speech data
No full textThe primary aim of this speech data set was to study the role of tone in the function word "ke" in the minimal pairs "ke motho" and in the function word "o" in "o batlang?" This data set is also useful for investigation of the vowel quality of the vowels containing in the speech recordings in terms of their frequency formants
NCHLT Afrikaans Auxiliary Speech Corpus
No full textThe corpus contains orthographically transcribed broadband speech in each of South Africa's eleven official languages. Transcriptions are provided in XML format
NCHLT isiZulu Auxiliary Speech Corpus
No full textThe corpus contains orthographically transcribed broadband speech in each of South Africa's eleven official languages. Transcriptions are provided in XML format
Sesotho vowel speech data set
No full textThe primary aim of this speech dataset was to collect a representative set of words in which all the Sesotho vowels are present. Some of them are supposedly produced at a high or a low tone. Researchers may consult standard works like that of Cole for the tone patterns as well as the quality of the vowels of these words. Recordings were made of ten Southern Sotho permanent inhabitants of the Phuthaditshaba region in the Eastern Free State, three of them male, ages unknown
Ex Machina: Using NLP and statistical learning models to model eyewitness statements and choosing behaviour
No full textThis curated database includes data from various of empirical studies where eyewitness statements and descriptions were collected. The original studies, including their datasets, the written document describing the original datasets and experiments, as well as the materials, are curated individually within this archive. The intention of this archive is to (1) allow other researchers access to a large archive of eyewitness statements and descriptions, and (2) regularly update the archive. The curated archive of eyewitness statements and descriptions are accompanied by R code that allows for analysis of the statements, where the eyewitness statements predict the eyewitness’ lineup decision
Sesotho tone data set
No full textThese recordings are of male and female speakers (11 for tasks 1 and 2; 10 for task 3) of the QwaQwa region (Eastern Free State). Ages of the speakers are not known, but all of them were at the time of the research either students or lecturers and administrative staff of the then University of QwaQwa. The dataset is divided into 3 subsets/tasks
NCHLT Xitsonga Auxiliary Speech Corpus
No full textThe corpus contains orthographically transcribed broadband speech in each of South Africa's eleven official languages. Transcriptions are provided in XML format
NCHLT Setswana Auxiliary Speech Corpus
No full textThe corpus contains orthographically transcribed broadband speech in each of South Africa's eleven official languages. Transcriptions are provided in XML format
NCHLT isiNdebele Auxiliary Speech Corpus
No full textThe corpus contains orthographically transcribed broadband speech in each of South Africa's eleven official languages. Transcriptions are provided in XML format
NCHLT Sesotho Auxiliary Speech Corpus
No full textThe corpus contains orthographically transcribed broadband speech in each of South Africa's eleven official languages. Transcriptions are provided in XML format