Discriminative Training for Speech Recognition is Compensating for Statistical Dependence on the HMM Framework
Title | Discriminative Training for Speech Recognition is Compensating for Statistical Dependence on the HMM Framework |
Publication Type | Conference Paper |
Year of Publication | 2012 |
Authors | Gillick, D., Wegmann S., & Gillick L. |
Page(s) | 4745-4748 |
Other Numbers | 3240 |
Abstract | The parameters of the standard Hidden Markov Model framework for speech recognition are typically trained via Maximum Likelihood. However, better recognition performance is achievable with discriminative training criteria like Maximum Mutual Information or Minimum Phone Error. While it is generally accepted that these discriminative criteria are better suited to minimizing Word Error Rate, there is very little qualitative intuition for how the improvements are achieved. Through a series of resampling experiments, we show that discriminative training (MPE in particular) appears to be compensating for a specific incorrect assumption of the HMM-that speech frames are conditionally independent. |
Acknowledgment | This work was partially supported by funding provided to ICSI through National Science Foundation grant IIS-1015930 (Exploratory Data Analysis for Speech Recognition). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors or originators and do not necessarily reflect the views of the National Science Foundation. |
URL | http://www.icsi.berkeley.edu/pubs/speech/gillickwegmannicassp12.pdf |
Bibliographic Notes | Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2012), pp. 4745-4748, Kyoto, Japan |
Abbreviated Authors | D. Gillick, S. Wegmann, and L. Gillick |
ICSI Research Group | Speech |
ICSI Publication Type | Article in conference proceedings |