NWU Institutional Repository

Constructing sub-word units for spoken term detection

Simple item page
dc.contributor.authorvan Heerden, Charl
dc.contributor.authorKarakos, Damianos
dc.contributor.authorNarasimhan, Karthik
dc.contributor.authorSchwartz, Richard
dc.contributor.authorDavel, Marelie H.
dc.date.accessioned2018-02-27T10:06:59Z
dc.date.available2018-02-27T10:06:59Z
dc.date.issued2017
dc.description.abstractSpoken term detection, especially of out-of-vocabulary (OOV) keywords, benefits from the use of sub-word systems. We experiment with different language-independent approaches to sub-word unit generation, generating both syllable-like and morpheme-like units, and demonstrate how the performance of syllable-like units can be improved by artificially increasing the number of unique units. The effect of unit choice is empirically evaluated using the eight languages from the 2016 IARPA BABEL evaluation. Index Terms— Spoken term detection, BABEL, sub-words, syllables, morphemes.en_US
dc.description.sponsorshipWe would like to thank all members of the Babelon team at Raytheon BBN Technologies, and especially Tanel Alumae, William Hartmann and Stavros Tsakalidis. This work was supported by the Intelligence Advanced Research Projects Activity (IARPA) via Department of Defense U.S. Army Research Laboratory contract number W911NF-12-C-0013. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon. Disclaimer: The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of IARPA, DoD/ARL, or the U.S. Government.en_US
dc.identifier.citationCharl van Heerden, Damianos Karakos, Karthik Narasimhan, Marelie Davel and Richard Schwartz, “Constructing sub-word units for spoken term detection”, in Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), pp 5780-5784, New Orleans, Louisiana, 2017. [http://engineering.nwu.ac.za/multilingual-speech-technologies-must/publications]
dc.identifier.urihttp://hdl.handle.net/10394/26442
dc.identifier.urihttps://pdfs.semanticscholar.org/f28e/39372d12d1279ea775a80df3e63f86069706.pdf
dc.language.isoenen_US
dc.publisherAcoustics, Speech and Signal Processing (ICASSP)en_US
dc.subjectSpoken term detection,en_US
dc.subjectEffect of unit choiceen_US
dc.subjectBABELen_US
dc.titleConstructing sub-word units for spoken term detectionen_US
dc.typePresentationen_US

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
vanheerden-2017-constructing-subword-units.pdf
Size:
177.46 KB
Format:
Adobe Portable Document Format
Description:
vanheerden-2017-constructing-subword-units
Download

License bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.61 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Faculty of Engineering