Using sub-word n-gram models for dealing with OOV in large vocabulary speech recognition for Latvian

Using sub-word n-gram models for dealing with OOV in large vocabulary speech recognition for Latvian Askars Salimbajevs author Jevgenijs Strigins author 2015-05 text Proceedings of the 20th Nordic Conference of Computational Linguistics (NODALIDA 2015) Beáta Megyesi editor Linköping University Electronic Press, Sweden Vilnius, Lithuania conference publication salimbajevs-strigins-2015-using https://aclanthology.org/W15-1837/ 2015-05 281 285