| 摘要 |
The purpose of the present invention is to provide a voice search technology which enables problem calls to be efficiently discovered and checked. Accordingly, a voice search system according to the present invention is provided with a call search database in which, for every reception channel and transmission channel, voice segment sequences in each of a plurality of pieces of recorded-call voice data are associated with prescribed keywords and time information, and stored. The voice search system searches the call search database on the basis of inputted search keywords, to acquire search-keyword-containing voice segment sequences. More specifically, in this voice search system, search-keyword-containing voice segment sequences, and occurrence times of said search-keyword-containing voice segment sequences are acquired as keyword search results, from among the plurality of pieces of recorded-call voice data, start times for the voice segment sequences of separate channels, said start times being immediately prior to the voice segment sequences acquired as the keyword search results, are acquired on the basis of the occurrence times in the keyword search results, and the start times are determined as voice-playback cue positions. These cue positions are outputted as voice search results. |
