Article (Published version) (352 Kb) - 
Free access
Highlights
Scientific Article
unige:75402 
 |
Bootstrapping a Statistical Speech Translator From a Rule-Based One |
|
| Authors | ||
| Published in | Linguistic Issues in Language Technology. 2012, vol. 5, no. 3 | |
| Abstract | We describe a series of experiments in which we start with English to French and English to Japanese versions of a rule-based speech translation system for a medical domain, and bootstrap corresponding statistical systems. Comparative evaluation reveals that the statistical systems are still slightly inferior to the rule-based ones, despite the fact that con- siderable effort has been invested in tuning both the recognition and translation components; however, a hybrid system is able to deliver a small but significant improvement in performance. In conclusion, we suggest that the hybrid architecture we describe potentially allows construction of limited-domain speech translation systems which combine substantial source-language coverage with high-precision translation. | |
| Keywords | Translation — Speech technology — Spoken language | |
| Full text | ||
| Structures | ||
| Citation (ISO format) | RAYNER, Emmanuel et al. Bootstrapping a Statistical Speech Translator From a Rule-Based One. In: Linguistic Issues in Language Technology, 2012, vol. 5, n° 3. https://archive-ouverte.unige.ch/unige:75402 |
314 hits 
50 downloads 
Contact an author
Update