This paper presents a new method of feature dimension reduction in hidden Markov modeling (HMM) for speech recognition. The key idea is to apply reduced rank maximum likelihood estimation in the M-step of the usual Baum-Welch algorithm for estimating HMM parameters such that the estimates of the Gaussian distribution parameters are restricted in a sub-space of reduced dimensionality. There are ...

The signal segmentation approach described herein assumes that the signal can be accurately modelled by a linear regression with piece-wise constant parameters. A simultaneous estimate of the change times is considered. The maximum likelihood and maximum a posteriori probability estimates are derived after marginalization of the linear regression parameters and the measurement noise variance, w...

Self-signaling models predict less selfish behavior in a probabilistic giving setting as individuals are expected to invest pro-social identity. However, there is also substantial evidence that people tend exploit situational excuses for choices (for instance, uncertainty) and behave more selfishly. We contrast these two motivations (identity management self-deception) experimentally order test...

This paper presents a new numerical approach for the full extraction of coupling-of-modes (COM) parameters by stationary and eigenfrequency analyses in finite element method (FEM). is fast extracting from results static analysis analysis. It avoids long calculation time admittance frequency response analysis, which commonly used COM parameters. In addition to usual (velocity, reflection coeffic...