摘要 |
PROBLEM TO BE SOLVED: To provide the voice summarizing device, which provides summarized voices that are made to be better understood by a user, by reproducing and outputting the summarized and important portion of the voices with a higher priority based on the summary data obtained by evaluating the time distribution of the voice feature group interval that are automatically extracted. SOLUTION: The device has an important interval extracting section (a feature extracting section 2→an important interval index storage section 4) which extracts an important interval from voices, a summary data processing means which is provided with a data processing section (a time portion evaluating section 5→a reproducing index storage section 6) that processes the important interval and obtains summary data and an outputting means (a voice waveform storage section 7→a voice outputting section 9) which outputs voices based on the summary data obtained by the summary data processing means. The section 4 evaluates the time distribution of the important interval and obtains the summary data which are determined based on the evaluation result. The outputting means reproduces and outputs the voices.
|