Jingle detection and identification in audio documents

This work addresses the soundtrack indexing of multime-dia documents. Our purpose is to detect and locate one or many jingles to structure the audio dataflow in program broadcasts (reports). Each jingle is commonly represented by a sequence of spectral vectors, considered as its “signa-ture”. Potential candidates are extracted from the data flow by computing an Euclidean distance. They are validated with heuristic rules. The system evaluation is performed on TV and radio corpora (more than 10 hours, 3 TV channels and 3 radio channels). First results show that the system is efficient: among 132 jingles to recognize, we have de-tected 130 with our reference jingle table of 32 different key sounds. 1