摘要 |
<p>A system and method for skimming digital audio (18) and video data (20) wherein the video data is partitioned into video segments.The method includes, selecting representative frames (64a, 64b, 64c, 64d) from each of the video segments, combining (235) the representative frames to form an assembled video sequence, identifying (230) keywords contained in a transcription of the audio data,extracting (237) portions of the audio data identified as keywords in the identifying step, assembling (239) an audio track in response to the extraction step, and outputting the video sequence in conjunction with the audio track.</p> |