ICPR
2006
IEEE
14 years 5 months ago
2006
IEEE
Speech recognition is usually based on Hidden Markov Models (HMMs), which represent the temporal dynamics of speech very efficiently, and Gaussian mixture models, which do non-opt...