16th Annual Conference of the International Speech Communication Association

Dresden, Germany
September 6-10, 2015

Exploring Minimal Pronunciation Modeling for Low Resource Languages

Marelie Davel (1), Etienne Barnard (1), Charl van Heerden (1), William Hartmann (2), Damianos Karakos (2), Richard Schwartz (2), Stavros Tsakalidis (2)

(1) North-West University, South Africa
(2) Raytheon BBN Technologies, USA

Pronunciation lexicons can range from fully graphemic (modeling each word using the orthography directly) to fully phonemic (first mapping each word to a phoneme string). Between these two options lies a continuum of modeling options. We analyze techniques that can improve the accuracy of a graphemic system without requiring significant effort to design or implement. The analysis is performed in the context of the IARPA Babel project, which aims to develop spoken term detection systems for previously unseen languages rapidly, and with minimal human effort. We consider techniques related to letter-to-sound mapping and language-independent syllabification of primarily graphemic systems, and discuss results obtained for six languages: Cebuano, Kazakh, Kurmanji Kurdish, Lithuanian, Telugu and Tok Pisin.

Full Paper

Bibliographic reference.  Davel, Marelie / Barnard, Etienne / Heerden, Charl van / Hartmann, William / Karakos, Damianos / Schwartz, Richard / Tsakalidis, Stavros (2015): "Exploring minimal pronunciation modeling for low resource languages", In INTERSPEECH-2015, 538-542.