| 摘要 |
<p>PROBLEM TO BE SOLVED: To provide a voice-document converting device which registers and retrieve XML document data without describing any special document for talk management in addition to the XML document data. SOLUTION: A voice XML conversion part 1 recognizes an input voice by using the grammar that a grammar generation part 5 automatically generates from an XML document stored in an XML document storage part 4. When the input indicates data registration, a data registration part 2 is selected and registers the XML document data to the XM document data stored in an XML document storage part 4. When the input indicates retrieval, on the other hand, a data retrieval part 3 is selected and retrives the XML document data stored in the XML document storage part 4. Then the retrieved XML document data are converted by a voice XML conversion part 1 into a document, which is outputted as a voice to a speaker or telephone line. The grammar for voice recognition can automatically be generated by using document type definitions(DTD) of the XM document data.</p> |
