Title :
On the first order equalization of hidden Markov models
Author :
Demirekler, Mübeccel
Author_Institution :
Dept. of Electr. & Electron. Eng., Middle East Tech. Univ., Ankara, Turkey
fDate :
6/1/1995 12:00:00 AM
Abstract :
The article exploits some properties of the “first order equalization” technique, used to increase the recognition performance of speech. It shows that if the eigenvalues of the covariance matrix of observations are restricted, then a solution exists for this problem, although it may not be unique,
Keywords :
Gaussian distribution; Gaussian processes; covariance matrices; eigenvalues and eigenfunctions; hidden Markov models; speech recognition; Gaussian distribution; covariance matrix; eigenvalues; first order equalization; hidden Markov models; observations; solution; speech recognition performance; Brain modeling; Electroencephalography; Frequency estimation; Hidden Markov models; Least squares approximation; Parameter estimation; Signal processing; Signal processing algorithms; Speech processing; Speech recognition;
Journal_Title :
Signal Processing, IEEE Transactions on