摘要:
The device comprises an image and/or sound sensing assembly (7), capable of retrieving at least a semantic element (41-43) from the content of audio-visual programmes being broadcast. It further comprises a unit identifying (8) said semantic elements, connected to a programme guide (20) including one or several chronologically arranged lists (21-23) of data sets respectively associated with the broadcast programmes. The identifying unit selects for at least one of the programmes being broadcast, the data set which is most probably correlated to said semantic elements. It carries out a prior learning session enabling to select the appropriate data sets from the semantic elements. A synthesizing unit (5) synchronises the audio-visual programmes with the selected data sets.