Speech recognizers typically use high-dimensional feature vectors to capture the essential cues for speech recognition purposes. The acoustics are then commonly modeled with a Hidden Markov Model with Gaussian Mixture Models as observation probability density functions. Using unrestricted Gaussian parameters might lead to intolerable model costs both evaluationand storagewise, which limits thei...

An approach is proposed for partial tying of states of tiedmixture hidden Markov models. To facilitate tying at the substate level, the state emission probabilities are constructed in two stages, or equivalently, are viewed as a ‘‘mixture of mixtures of Gaussians.’’ This paradigm allows, and is complemented with, an optimization technique to seek the best complexity-accuracy tradeoff solution, ...

The objective of this tutorial is to introduce basic concepts of a Hidden Markov Model (HMM). The tutorial is intended for the practicing engineer, biologist, linguist or programmer who would like to learn more about the above mentioned fascinating mathematical models and include them into one’s repertoire. This part of the tutorial is devoted to the basic concepts of a Hidden Markov Model. You...

We compare four different approaches towards modeling frame-level errors in GSM channels. One of these, the Markov-based Trace Analysis model (MTA), was developed for the purpose of modeling a GSM channel. The next two, -th-order Markov models and hidden Markov models (HMMs) have been widely used to model loss in wired networks. All three of these have difficulty modeling empirical GSM framelev...

We address a method to efficiently select Gaussian mixtures for fast acoustic likelihood computation. It makes use of context-independent models for selection and back-off of corresponding triphone models. Specifically, for the kbest phone models by the preliminary evaluation, triphone models of higher resolution are applied, and others are assigned likelihoods with the monophone models. This s...

In this paper, large vocabulary children’s speech recognition is investigated by using the Deep Neural Network Hidden Markov Model (DNN-HMM) hybrid and the Subspace Gaussian Mixture Model (SGMM) acoustic modeling approach. In the investigated scenario training data is limited to about 7 hours of speech from children in the age range 7-13 and testing data consists in read clean speech from child...

This report explains the theory of Hidden Markov Models (HMMs). The emphasis is on the theory aspects in conjunction with the implementation issues that are encountered in a floating point processor. The main theory and implementation issues are based on the use of a Gaussian Mixture Model (GMM) as the state density in the HMM, and a Continuous Density Hidden Markov Model (CDHMM) is assumed. Su...

In this paper, we first introduce the use of Gaussian mixture models (GMM) for Chinese tone classification in continuous speech. Then, we explain how to integrate it with the HMM-based speech recognition system. Finally, we provide the tone classification accuracy of this probabilistic method which is tested with Chinese continuous speech database of national “863” project.