Reducing errors by increasing the error rate: MLP Acoustic Modeling for Broadcast News Transcription

Morgan, Nelson; Ellis, Daniel P. W.; Fosler-Lussier, Eric; Janin, Adam; Kingsbury, Brian

We describe some aspects of a Broadcast News recognition system based on hybrid HMM/MLP acoustic modeling. These include the use of novel 'modulation spectrogram' features which are combined with conventional models at the posterior probability level, some experiments with nonlinear segment normalization, and an investigation of the interaction of model size and training set size for an multilayer perceptron (MLP) acoustic classifier. We also report preliminary results of incorporating gender-dependence into this system.


Also Published In

Proceedings of the DARPA Broadcast News Workshop, February 28-March 3, 1999, Hilton at Washington Dulles Airport, Herndon, Virginia
Information Technology Laboratory, National Institute of Standards and Technology

More About This Work

Academic Units
Electrical Engineering
Published Here
July 3, 2012