no code implementations • LREC 2012 • Igor Odriozola, Eva Navas, Inma Hernaez, I{\~n}aki Sainz, Ibon Saratxaga, Jon S{\'a}nchez, Daniel Erro
The distribution of the scores for erroneous pronunciation has been calculated inserting controlled errors in the dictionary, so that each changed phoneme has been randomly replaced by a phoneme from the same group.