Combining Knowledge Sources to Reorder N-Best Speech Hypothesis Lists

A simple and general method is described that can combine different knowledge sources to reorder N-best lists of hypotheses produced by a speech recognizer. The method is automatically trainable, acquiring information from both positive and negative examples. In experiments, the method was tested on a 1000-utterance sample of unseen ATIS data.

• Type: Conference
• Year: 1994
• Where: NAACL
• Authors: Manny Raynez, David Carter, Vassilios Digalakis, Patti Price
• Keywords: General Method, NAACL 1994, NAACL 2007, Speech Recognizer, Unseen Atis Data