Audio keyword extraction by unsupervised word discovery

Armando Muscariello (1) , Guillaume Gravier (1) , Frédéric Bimbot (1)

1 METISS - Speech and sound data modeling and processing

In real audio data, frequently occurring patterns often convey relevant information on the overall content of the data. The possibility to extract meaningful portions of the main content by identifying such key patterns, can be exploited for providing audio summaries and speeding up the access to relevant parts of the data. We refer to these patterns as audio motifs in analogy with the nomenclature in its counterpart task in biology. We describe a framework for the discovery of audio motifs in streams in an unsupervised fashion, as no acoustic or linguistic models are used. We define the fundamental problem by decomposing the overall task into elementary subtasks; then we propose a solution that combines a one-pass strategy that exploits the local repetitiveness of motifs and a dynamic programming technique to detect repetitions in audio streams. Results of an experiment on a radio broadcast show are shown to illustrate the effectiveness of the technique in providing audio summaries of real data.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Armando Muscariello, Guillaume Gravier, Frédéric Bimbot. Audio keyword extraction by unsupervised word discovery. INTERSPEECH 2009: 10th Annual Conference of the International Speech Communication Association, Sep 2009, Brighton, United Kingdom. ⟨inria-00551769⟩