Name pronunciation in German text-to-speech synthesis (1997)
Cached
Download Links
| Venue: | IN PROC. 5TH CONF. ON APPLIED NATURAL LANGUAGE PROCESSING |
| Citations: | 5 - 2 self |
BibTeX
@INPROCEEDINGS{Jannedy97namepronunciation,
author = {Stefanie Jannedy and Bernd Möbius},
title = {Name pronunciation in German text-to-speech synthesis},
booktitle = {IN PROC. 5TH CONF. ON APPLIED NATURAL LANGUAGE PROCESSING},
year = {1997},
pages = {49--56},
publisher = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We describe the name analysis and pronunciation component in the German version of the Bell Labs multilingual text-tospeech system. We concentrate on street names because they encompass interest- ing aspects of geographical and personal names. The system was implemented in the framework of finite-state transducer technology, using linguistic criteria as well as frequency distributions derived from a database. In evaluation experiments, we compared the performances of the generalpurpose text analysis and the name-specific system on training and test materials. The name-specific system significantly outperforms the generic system. The error rates compare favorably with results reported in the research literature. Finally, we discuss areas for future work.
Keyphrases
name pronunciation german text-to-speech synthesis name-specific system street name future work evaluation experiment frequency distribution pronunciation component german version personal name generic system bell lab multilingual text-tospeech system name analysis linguistic criterion interest ing aspect error rate generalpurpose text analysis finite-state transducer technology test material research literature
