Copyright © 2006 The Institute of Electronics, Information and Communication Engineers
Special Section on Statistical Modeling for Speech Processing -- Papers |
What HMMs Can Do
The author is with the Dept. of Electrical Engineering, the University of Washington, Seattle, Box 352500, Seattle WA, 98195–2500, U.S.A. E-mail: bilmes{at}ee.washington.edu
Since their inception almost fifty years ago, hidden Markov models (HMMs) have have become the predominant methodology for automatic speech recognition (ASR) systems—today, most state-of-the-art speech systems are HMM-based. There have been a number of ways to explain HMMs and to list their capabilities, each of these ways having both advantages and disadvantages. In an effort to better understand what HMMs can do, this tutorial article analyzes HMMs by exploring a definition of HMMs in terms of random variables and conditional independence assumptions. We prefer this definition as it allows us to reason more throughly about the capabilities of HMMs. In particular, it is possible to deduce that there are, in theory at least, no limitations to the class of probability distributions representable by HMMs. This paper concludes that, in search of a model to supersede the HMM (say for ASR), rather than trying to correct for HMM limitations in the general case, new models should be found based on their potential for better parsimony, computational requirements, and noise insensitivity.
Key Words: automatic speech recognition, hidden Markov models, HMMs, time-series processes, hand-writing recognition, graphical models, dynamic Bayesian networks, dynamic graphical models, stochastic processes, time-series densities, bio-informatics
Manuscript received October 3, 2005. Manuscript revised December 20, 2005.