Modeling Duration in a Hidden Markov Model with the Exponential Family
نویسنده
چکیده
Explicit duration modeling has been shown to increase the eeectiveness of hidden Markov models in automatic speech recognition. Ferguson found the optimum parameters of the duration model for the case where duration is assumed to be distributed according to a non-parametric probability mass function. Levinson determined the best gamma density to model duration. In this paper, duration is assumed to be modeled by some probability mass function in the exponential family. An iterative procedure for determining the maximum likelihood parameters is presented. Also given is a method for choosing an appropriate member from the exponential family.
منابع مشابه
Improving Phoneme Sequence Recognition using Phoneme Duration Information in DNN-HSMM
Improving phoneme recognition has attracted the attention of many researchers due to its applications in various fields of speech processing. Recent research achievements show that using deep neural network (DNN) in speech recognition systems significantly improves the performance of these systems. There are two phases in DNN-based phoneme recognition systems including training and testing. Mos...
متن کاملMarkov Chain Analogue Year Daily Rainfall Model and Pricing of Rainfall Derivatives
In this study we model the daily rainfall occurrence using Markov Chain Analogue Yearmodel (MCAYM) and the intensity or amount of daily rainfall using three different probability distributions; gamma, exponential and mixed exponential distributions. Combining the occurrence and intensity model we obtain Markov Chain Analogue Year gamma model (MCAYGM), Markov Chain Analogue Year exponentia...
متن کاملمدل سازی فضایی-زمانی وقوع و مقدار بارش زمستانه در گستره ایران با استفاده از مدل مارکف پنهان
Multi site modeling of rainfall is one of the most important issues in environmental sciences especially in watershed management. For this purpose, different statistical models have been developed which involve spatial approaches in simulation and modeling of daily rainfall values. The hidden Markov is one of the multi-site daily rainfall models which in addition to simulation of daily rainfall...
متن کاملEfficient duration and hierarchical modeling for human activity recognition
a r t i c l e i n f o a b s t r a c t A challenge in building pervasive and smart spaces is to learn and recognize human activities of daily living (ADLs). In this paper, we address this problem and argue that in dealing with ADLs, it is beneficial to exploit both their typical duration patterns and inherent hierarchical structures. We exploit efficient duration modeling using the novel Coxian ...
متن کاملA Hidden Markov Model for Modeling Pavement Deterioration under Incomplete Monitoring Data
In this paper, the potential use of an exponential hidden Markov model to model a hidden pavement deterioration process, i.e. one that is not directly measurable, is investigated. It is assumed that the evolution of the physical condition, which is the hidden process, and the evolution of the values of pavement distress indicators, can be adequately described using discrete condition states and...
متن کامل