Segmentation, Classification and Clustering of an Italian Broadcast News Corpus
Mauro Cettolo · 2000
This work reports on preliminary activity at ITC-irst on the problem of acoustic segmentation, classification and clustering of an Italian audio broadcast news corpus. The approach is based on the following stages. First, the input data stream is segmented by detecting spectral changes through the Bayesian Information Criterion (BIC). Second, segments are classified in terms of acoustic conditions, modeled by mixtures of Gaussians. Finally, segments from the same speakers are clustered, by using again the BIC. The scheme proposed for the automatic segmentation, classification and clustering causes a degradation of the recognition error rate, with respect to the fully supervised experiment, equal to 1.3% before adaptation, and 3.4% after adaptation. Introduction Automatic segmentation, classification and clustering in terms of acoustic source and acoustic channel of an audio stream is a challenging problem to which many research groups have recently devoted attention (Chen & G...