DIALOG RECORDER AND DIALOG RECORDING METHOD

PROBLEM TO BE SOLVED: To provide a dialog recorder for recording/reproducing voice of a specified speaker and voice of other speakers by distinguishing them. SOLUTION: Voice of dialog is fetched by a voice input part 1, its waveform data are transferred to an encoding part 2 and encoded to information in a format which can be treated by a computer. The encoded information is transmitted to an utterance person specification part 4 and whether or not an utterance person is registered is judged. In the case of utterance by a non-registrant, a recognition control part 5 stores its voice information in a storage part 8 as it is and on the other hand, in the case of utterance by a registrant, the voice information is handed over to a voice recognition part 6. The voice recognition part 6 converts the voice information into text information and returns it to the recognition control part 5 by utilizing a featured dictionary of the utterance of the registrant stored in a feature storage part 7. Then, the recognition control part 5 stores the encoded voice information in the storage part 8 with the text information. COPYRIGHT: (C)2005,JPO&NCIPI