Peer-reviewed research

A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition

Lawrence R. Rabiner, Proceedings of the IEEE, vol. 77, no. 2, pp. 257-286, 1989 · DOI 10.1109/5.18626

Read on October 17, 2026

What it is: The standard tutorial on hidden Markov models (HMMs), the statistical model that powered most speech recognizers from the 1980s to the early 2010s. It states the three basic problems that must be solved before an HMM can be used on speech.
How much weight it carries: Peer-reviewed, in the Proceedings of the IEEE, and the most cited introduction to the topic. It is a 1989 text: it describes the theory and the systems of its time, not current neural models.

Facts taken from this source

Rabiner's 1989 tutorial says HMMs must solve three fundamental problems before they apply to speech: evaluating the likelihood of an observation sequence, finding the best sequence of model states, and adjusting model parameters to fit the observed signal.
“evaluation of the probability (or likelihood) of a sequence of observations given a specific HMM; the determination of a best sequence of model states; and the adjustment of model parameters so as to best account for the observed signal” (Introduction)