Integrating recognition and retrieval with user feedback: A new framework for spoken term detection

People usually consider recognition and retrieval as two cascaded independent modules for spoken term detection. Retrieval techniques were assumed to be applied on top of some ASR output, with performance depending on ASR accuracy. In this paper, we propose a new framework: to integrate the two parts into a single task. This can be achieved by adjusting the acoustic model parameters, borrowing the principle of Minimum Classification Error (MCE), based on user feedback. The modified acoustic models then give updated posterior probabilities for the lattice-based structures used in spoken term detection. Encouraging results were obtained on a bilingual course lecture corpus in preliminary experiments.