An introduction to hidden Markov models

By Lawrence R. Rabiner, Biing-Hwang Fred Juang. 1986

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Survey on Neural Speech Synthesis (2021, methodology): “Compared to previous concatenative speech synthesis, HMMbased parametric synthesis is more flexible in changing speaker identities, emotions, and speaking styles [350].”
Bernstein's inequalities for general Markov chains (2018, unknown): “…frequently used in applications such as numerical integration via Monte Carlo methods (Gilks et al., 1995), Hidden Markov Models (Baum et al., 1970; Rabiner, 1989; Bishop, 2006), and optimization algorithms like stochastic gradient descent (Bottou, 2010; Goodfellow et al., 2016) and simulated…”
Recognizing teleoperated manipulations (1993, methodology): “The HMM is initialized with a directed graph (number of states and edge topology) and probability distributions for the start states, transitions and obser- grasp carry press slide none 0 5 10 15 20 25 PASS 1 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +”
Automatic Signature Verification: The State of the Art (2008, background): “A comprehensive discussion on HMM is beyond the aim of this paper and can be found in the literature [271].”
Conditional Heteroskedasticity Driven by Hidden Markov Chains (2001, methodology): “These Markov-switching models clearly have many common features with the hidden Markov models introduced by Baum and Petrie (1966) and successfully used in the ®eld of speech pattern modeling (see for instance Rabiner and Juang, 1986; Elliot et al., 1995).”
Age and gender classification using convolutional neural networks (2015, methodology): “Finally, instead of GMM, Hidden-MarkovModel, super-vectors [40] were used in [56] for representing face patch distributions.”
Video-based event recognition: activity representation and probabilistic recognition methods (2004, methodology): “A more accurate way of computing PðSiðti ;tÞ jOðti ;tÞÞ is to model the event duration explicitly (also known as a semi-HMM) [25].”
MASSACHUSETTS INSTITUTE OF TECHNOLOGY ARTIFICIAL INTELLIGENCE LABORATORY and CENTER FOR BIOLOGICAL AND COMPUTATIONAL LEARNING DEPARTMENT OF BRAIN AND COGNITIVE SCIENCES (1996, unknown): “Due to the simplicity and e ciency of its parameter estimation algorithm, the hidden Markov model (HMM) has emerged as one of the basic statistical tools for modeling discrete time series, nding widespread application in the areas of speech recognition (Rabiner and Juang, 1986) and computational mol”
Deep Temporal Sigmoid Belief Networks for Sequence Modeling (2015, background): “The multi-layered model is designed by constructing a hierarchy of temporal sigmoid belief networks (TSBNs), defined as a sequential stack of sigmoid belief networks (SBNs).”
Learning API Usages from Bytecode: A Statistical Approach (2015, methodology): “In general, Baum-Welch algorithm is often used to estimate parameters of Hidden Markov Model [34].”
Data Mining for Wearable Sensors in Health Monitoring Systems: A Review of Recent Trends and Challenges (2013, background): “Hidden Markov model (HMM) is a probabilistic model which is convenient for modeling sequential data [78].”
Offline Arabic handwriting recognition: a survey (2006, background): “Hidden Markov models (HMMs) are also appropriate for learning characteristics that are difficult to describe intuitively [13].”
A probabilistic approach for damage identification and crack mode classification in reinforced concrete structures (2013, background): “…Pr (ij~xt, lk)3(~xt ~mk + 1i )(~xt ~mk + 1i ) T PT t = 1 Pr (ij~xt, lk) ð10Þ The use of a GMM may be motivated by the intuitive notion that the individual component densities may model some underlying set of hidden classes (Rabiner and Juang, 1986; Rammohan and Taha, 2005; Reynolds and Rose, 1995).”
A Unifying Review of Linear Gaussian Models (1999, methodology): “The EM algorithm for linear gaussian dynamical systems was originally derived by Shumway and Stoffer (1982) and recently reintroduced (and extended) in the neural computation field by Ghahramani and Hinton (1996a,b).”
Robotics-Based Location Sensing Using Wireless Ethernet (2002, methodology): “We model a moving operator trying to track her position as a hidden Markov model (HMM) [38, 41 ].”
Introduction to Digital Speech Processing (2007, methodology): “The most widely used method of building acoustic models (for both phonemes and words) is the use of a statistical characterization known as Hidden Markov Models (HMMs) [33, 69, 97, 98].”