Robinson A. J. author Cook G. D. author Ellis Daniel P. W. author Columbia University. Electrical Engineering Fosler-Lussier Eric author Renals S. J. author Williams D. A. G. author Columbia University. Electrical Engineering originator text Articles 2002 manuscript version English This paper describes connectionist techniques for recognition of Broadcast News. The fundamental difference between connectionist systems and more conventional mixture-of-Gaussian systems is that connectionist models directly estimate posterior probabilities as opposed to likelihoods. Access to posterior probabilities has enabled us to develop a number of novel approaches to confidence estimation, pronunciation modelling and search. In addition we have investigated a new feature extraction technique based on the modulation-filtered spectrogram (MSG), and methods for combining multiple information sources. We have incorporated all of these techniques into a system for the transcription of Broadcast News, and we present results on the 1998 DARPA Hub-4E Broadcast News evaluation data. Artificial intelligence Communication 37 1-2 27 45 2002-05 http://dx.doi.org/10.1016/S0167-6393(01)00058-9 http://hdl.handle.net/10022/AC:P:12568 NNC NNC 2012-02-15 16:02:25 -0500 2013-02-11 14:25:46 -0500 6573 eng