| 摘要 |
A portion of audio content of a multimedia program, such as a television program, is captured from a network. An audio fingerprint is generated based on the portion of audio content, and the audio fingerprint is matched to one of multiple theme song fingerprints stored in a database. An expected theme song time offset associated with the matched theme song fingerprint is retrieved from the database. It is determined whether the program is running on-schedule, based on the time the portion of audio content occurred, a scheduled start time of the program, and/or the expected theme song time offset. If it is determined that the program is running off-schedule, an adjusted start time and/or an adjusted end time of the program are calculated. The program is recorded by a recorder based on the adjusted start time and/or the adjusted end time.
|
