The authors describe a new language-independent technique for automatically identifying errors in an electronic pronunciation dictionary by analyzing the source of conflicting patterns directly.They evaluate the effectiveness of the technique in two ways: they perform a controlled experiment using artificially corrupted data (allowing us to measure precision and recall exactly); and then apply the technique to a real-world pronunciation dictionary, demonstrating its effectiveness in practice. They also introduce a new freely available pronunciation resource (the RCRL Afrikaans Pronunciation Dictionary), the largest such dictionary that currently exists.
Reference:
Davel, MH and De Wet, F. 2010. Verifying pronunciation dictionaries using conflict analysis. Proceedings of the 11th Annual Conference of the International Speech Communication Association (INTERSPEECH 2010), Makuhari, Japan, 26-30 September 2010, pp 1898-1901
Davel, M., & De Wet, F. (2010). Verifying pronunciation dictionaries using conflict analysis. http://hdl.handle.net/10204/4693
Davel, MH, and Febe De Wet. "Verifying pronunciation dictionaries using conflict analysis." (2010): http://hdl.handle.net/10204/4693
Davel M, De Wet F, Verifying pronunciation dictionaries using conflict analysis; 2010. http://hdl.handle.net/10204/4693 .
Proceedings of the 11th Annual Conference of the International Speech Communication Association (INTERSPEECH 2010), Makuhari, Japan, 26-30 September 2010