Difference between revisions of "Mining the Spoken Wikipedia for Speech Data and Beyond"
(Links) |
(Infobox work) |
||
| Line 1: | Line 1: | ||
| + | {{Infobox work | ||
| + | | title = Mining the Spoken Wikipedia for Speech Data and Beyond | ||
| + | | date = 2016 | ||
| + | | authors = [[Arne Köhn]]<br />[[Florian Stegen]]<br />[[Timo Baumann]] | ||
| + | | link = http://edoc.sub.uni-hamburg.de/informatik/volltexte/2016/220/pdf/koehn_spoken_wikipedia.pdf | ||
| + | }} | ||
'''Mining the Spoken Wikipedia for Speech Data and Beyond''' - scientific work related to [[Wikipedia quality]] published in 2016, written by [[Arne Köhn]], [[Florian Stegen]] and [[Timo Baumann]]. | '''Mining the Spoken Wikipedia for Speech Data and Beyond''' - scientific work related to [[Wikipedia quality]] published in 2016, written by [[Arne Köhn]], [[Florian Stegen]] and [[Timo Baumann]]. | ||
== Overview == | == Overview == | ||
Authors present a corpus of time-aligned spoken data of [[Wikipedia]] articles as well as the pipeline that allows to generate such corpora for many languages. There are initiatives to create and sustain spoken Wikipedia versions in many languages and hence the data is freely available, grows over time, and can be used for automatic corpus creation. Authors pipeline automatically downloads and aligns this data. The resulting German corpus currently totals 293h of audio, of which authors align 71h in full sentences and another 86h of sentences with some missing words. The English corpus consists of 287h, for which authors align 27h in full sentence and 157h with some missing words. Results are publically available. | Authors present a corpus of time-aligned spoken data of [[Wikipedia]] articles as well as the pipeline that allows to generate such corpora for many languages. There are initiatives to create and sustain spoken Wikipedia versions in many languages and hence the data is freely available, grows over time, and can be used for automatic corpus creation. Authors pipeline automatically downloads and aligns this data. The resulting German corpus currently totals 293h of audio, of which authors align 71h in full sentences and another 86h of sentences with some missing words. The English corpus consists of 287h, for which authors align 27h in full sentence and 157h with some missing words. Results are publically available. | ||
Revision as of 09:22, 15 April 2020
Authors | Arne Köhn Florian Stegen Timo Baumann |
|---|---|
Publication date | 2016 |
Links | Original |
Mining the Spoken Wikipedia for Speech Data and Beyond - scientific work related to Wikipedia quality published in 2016, written by Arne Köhn, Florian Stegen and Timo Baumann.
Overview
Authors present a corpus of time-aligned spoken data of Wikipedia articles as well as the pipeline that allows to generate such corpora for many languages. There are initiatives to create and sustain spoken Wikipedia versions in many languages and hence the data is freely available, grows over time, and can be used for automatic corpus creation. Authors pipeline automatically downloads and aligns this data. The resulting German corpus currently totals 293h of audio, of which authors align 71h in full sentences and another 86h of sentences with some missing words. The English corpus consists of 287h, for which authors align 27h in full sentence and 157h with some missing words. Results are publically available.
