13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Assessing Agreement Level Between Forced Alignment Models With Data From Endangered Language Documentation Corpora

Christian T. DiCanio (1), Hosung Nam (1), Douglas H. Whalen (1,2,3), H. Timothy Bunnell (4,5), Jonathan D. Amith (6,7), Rey Castillo Garcia (8)

(1) Haskins Laboratories, New Haven, CT, USA
(2) Speech-Language-Hearing Program, CUNY Graduate Center, New York, NY, USA
(3) Endangered Language Fund, New Haven, CT, USA
(4) Nemours Biomedical Research, Alfred I. duPont Hospital for Children, Wilmington, DE, USA
(5) Department of Computer and Information Sciences, University of Delaware, Newark, DE, USA
(6) Department of Anthropology, Gettysburg College, Gettysburg, Pennsylvania, USA
(7) Smithsonian Institution, Washington D.C., USA
(8) CIESAS, Mexico City, D.F., Mexico

Automatic forced alignment between transcriptions has achieved high levels of agreement for languages with large corpora, but the technique holds great promise for work on all languages. Here, we apply two forced alignment programs to data from an endangered Mixtecan language of Mexico. Both yielded a majority of boundaries within 20 ms of hand-labeled ones. Phonemes with fairly steady-state elements (e.g. nasals, fricatives) were more accurately labeled than others. Forced alignment thus may increase efficiency of labeling texts from smaller languages, at least in cases where the phoneme inventories are similar to those of the languages of the training.

Index Terms: speech recognition, phonetics, linguistics

Full Paper

Bibliographic reference.  DiCanio, Christian T. / Nam, Hosung / Whalen, Douglas H. / Bunnell, H. Timothy / Amith, Jonathan D. / Castillo Garcia, Rey (2012): "Assessing agreement level between forced alignment models with data from endangered language documentation corpora", In INTERSPEECH-2012, 130-133.