70
Voted
NAACL
2001
15 years 3 days ago
2001
We present a simple architecture for parsing transcribed speech in which an edited-word detector first removes such words from the sentence string, and then a standard statistical...