| 摘要 |
PROBLEM TO BE SOLVED: To efficiently obtain a target retrieval result in a method of extracting a keyword from a result of speech recognition of a speech of a customer or a staff answering to the customer and performing retrieval from a knowledge base based upon the keyword. SOLUTION: The speech recognition, keyword extraction, and knowledge base retrieval coordinating device includes a speech recognition unit which speech recognizes an utterance, a keyword extraction unit which extracts a keyword from a recognition result generated by the speech recognition unit, and a knowledge base retrieval unit which performs retrieval using as a retrieval key the keyword extracted by the keyword extraction unit. Different words are registered which mean the same object among entries into three kinds of language processing mechanisms for indexes of a speech recognition dictionary, an extraction object keyword list, and knowledge base, and the correspondence among them is managed. COPYRIGHT: (C)2008,JPO&INPIT
|
