Computer personnel research - Issues and progress in the 60's
David B. Mayer, Ashford W. Stalnaker
ACM SIGMIS CPR 1967
We employ a variant of the popular Adaboost algorithm to train multiple acoustic models such that the aggregate system exhibits improved performance over the individual recognizers. Each model is trained sequentially on re-weighted versions of the training data. At each iteration, the weights are decreased for the frames that are correctly decoded by the current system. These weights are then multiplied with the frame-level statistics for the decision trees and Gaussian mixture components of the next iteration system. The composite system uses a log-linear combination of HMM state observation likelihoods. We report experimental results on several broadcast news transcription setups which differ in the language being spoken (English and Arabic) and amounts of training data. Additionally, we study the impact of boosting on maximum likelihood (ML) and discriminatively trained acoustic models. Our findings suggest that significant gains can be obtained for small amounts of training data even after feature and model-space discriminative training. © 2011 Elsevier B.V. All rights reserved.
David B. Mayer, Ashford W. Stalnaker
ACM SIGMIS CPR 1967
Ken C.L. Wong, Satyananda Kashyap, et al.
Pattern Recognition Letters
Opher Etzion
DEBS 2007
Cen Rao, Alexei Gritai, et al.
ICCV 2003