Learning Bayesian networks for semantic frame composition in a spoken dialog system

International audience A stochastic approach based on Dynamic Bayesian Networks (DBNs) is introduced for spoken language understanding. DBN-based models allow to infer and then to compose semantic frame-based tree structures from speech transcriptions. Experimental results on the French MEDIA dialog corpus show the appropriateness of the technique which both lead to good tree identification results and can provide the dialog system with n-best lists of scored hypotheses.