We propose a new blind segmentation approach to acoustic event detection (AED) based on i-vectors. Conventional approaches to AED often required well-segmented data with non-overlapping boundaries for competing events. Inspired by block-based automatic image annotation in image retrieval tasks, we blindly segment audio streams into equal-length pieces, label the underlying observed acoustic events with multiple categories and with no event boundary information, extract i-vector for them, and perform classification using support vector machine and maximal figureof- merit based classifiers. Experiments on various sets of audio data show promising results with an average of 8% absolute gain in F1 over the conventional hidden Markov model based approach. An enhanced robustness at different noise levels is also observed. The key to the success lies in the enhanced discrimination power offered by the i-vector representation of the acoustic data.
Bibliographic reference. Huang, Zhen / Cheng, You-Chi / Li, Kehuang / Hautamäki, Ville / Lee, Chin-Hui (2013): "A blind segmentation approach to acoustic event detection based on i-vector", In INTERSPEECH-2013, 2282-2286.