dc.contributor.author |
Badenhorst, J
|
|
dc.contributor.author |
Van Heerden, C
|
|
dc.contributor.author |
Davel, M
|
|
dc.contributor.author |
Barnard, E
|
|
dc.date.accessioned |
2012-04-16T15:44:28Z |
|
dc.date.available |
2012-04-16T15:44:28Z |
|
dc.date.issued |
2011-08 |
|
dc.identifier.citation |
Badenhorst, J, Van Heerden, C, Davel, M and Barnard, E. 2011. Collecting and evaluating speech recognition corpora for 11 South African languages. Language Resources and Evaluation, vol. 45(3), pp 289-309 |
en_US |
dc.identifier.issn |
1574-020X |
|
dc.identifier.issn |
1574-0218 |
|
dc.identifier.uri |
http://www.springerlink.com/content/m772051343jg875k/fulltext.pdf
|
|
dc.identifier.uri |
http://hdl.handle.net/10204/5770
|
|
dc.description |
Copyright: 2011 Springer-Verlag. This is the pre-print version of the work. The definitive version is published in Language Resources and Evaluation, vol. 45(3), pp 289-309 |
en_US |
dc.description.abstract |
The authors describe the Lwazi corpus for automatic speech recognition (ASR), a new telephone speech corpus which contains data from the eleven official languages of South Africa. Because of practical constraints, the amount of speech per language is relatively small compared to major corpora in world languages, and they report on their investigation of the stability of the ASR models derived from the corpus. They also report on phoneme distance measures across languages, and describe initial phone recognisers that were developed using this data. They find that a surprisingly small number of speakers (fewer than 50) and around 10 to 20 hours of speech per language are sufficient for the purposes of acceptable phone-based recognition. |
en_US |
dc.language.iso |
en |
en_US |
dc.publisher |
Springer Science+Business Media B.V. |
en_US |
dc.relation.ispartofseries |
Workflow;8149 |
|
dc.subject |
South Afrcan languages |
en_US |
dc.subject |
Lwazi corpus |
en_US |
dc.subject |
Automatic speech recognition (ASR) |
en_US |
dc.subject |
Speech recognition |
en_US |
dc.subject |
Speech recognition evaluation |
en_US |
dc.title |
Collecting and evaluating speech recognition corpora for 11 South African languages |
en_US |
dc.type |
Article |
en_US |
dc.identifier.apacitation |
Badenhorst, J., Van Heerden, C., Davel, M., & Barnard, E. (2011). Collecting and evaluating speech recognition corpora for 11 South African languages. http://hdl.handle.net/10204/5770 |
en_ZA |
dc.identifier.chicagocitation |
Badenhorst, J, C Van Heerden, M Davel, and E Barnard "Collecting and evaluating speech recognition corpora for 11 South African languages." (2011) http://hdl.handle.net/10204/5770 |
en_ZA |
dc.identifier.vancouvercitation |
Badenhorst J, Van Heerden C, Davel M, Barnard E. Collecting and evaluating speech recognition corpora for 11 South African languages. 2011; http://hdl.handle.net/10204/5770. |
en_ZA |
dc.identifier.ris |
TY - Article
AU - Badenhorst, J
AU - Van Heerden, C
AU - Davel, M
AU - Barnard, E
AB - The authors describe the Lwazi corpus for automatic speech recognition (ASR), a new telephone speech corpus which contains data from the eleven official languages of South Africa. Because of practical constraints, the amount of speech per language is relatively small compared to major corpora in world languages, and they report on their investigation of the stability of the ASR models derived from the corpus. They also report on phoneme distance measures across languages, and describe initial phone recognisers that were developed using this data. They find that a surprisingly small number of speakers (fewer than 50) and around 10 to 20 hours of speech per language are sufficient for the purposes of acceptable phone-based recognition.
DA - 2011-08
DB - ResearchSpace
DP - CSIR
KW - South Afrcan languages
KW - Lwazi corpus
KW - Automatic speech recognition (ASR)
KW - Speech recognition
KW - Speech recognition evaluation
LK - https://researchspace.csir.co.za
PY - 2011
SM - 1574-020X
SM - 1574-0218
T1 - Collecting and evaluating speech recognition corpora for 11 South African languages
TI - Collecting and evaluating speech recognition corpora for 11 South African languages
UR - http://hdl.handle.net/10204/5770
ER -
|
en_ZA |