Petr Plecháč :: Corpus of Czech Verse and Beyond ASEEES 2016
Versification RG, ICL CAS | Institute of the Czech National Corpus

Detection of rhymes

Learning

Split final words into relevant components (vowels and consonant clusters), eg.:
láska [ l a: s k a ] => [a:]c4 [sk]c3 [a]c2 [∅]c1
maska [ m a s k a ] => [a]c4 [sk]c3 [a]c2 [∅]c1

Probability of láska :: maska being rhyme

p1(c4) ... probability of [a:] meeting [a] at c4 in training set

p0(c4) ... probability of [a:] meeting [a] at c4 in entire corpus

...