Sven E. Krüger

Mixture of Support Vector Machines for HMM based Speech Recognition

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-optimally the classification of speech into single speech units (phonemes). In this paper we use parallel mixtures of support vector machines (SVMs) for classification by integrating this method in a HMM-based speech recognition system. SVMs are very appealing due to their association with statistical learning theory and have already shown good results in pattern recognition and in continuous speech recognition. They suffer however from the effort for training which scales at least quadratic with respect to the number of training vectors. The SVM mixtures need only nearly linear training time making it easier to deal with the large amount of speech data. In our hybrid system we use the SVM mixtures as acoustic models in a HMM-based decoder. We train and test the hybrid system on the DARPA resource management (RM1) corpus, showing better performance than HMM-based decoder using Gaussian mixtures

Kernel least-squares models using updates of the pseudoinverse

Sparse nonlinear classification and regression models in reproducing kernel Hilbert spaces (RKHSs) are considered. The use of Mercer kernels and the square loss function gives rise to an overdetermined linear least-squares problem in the corresponding RKHS. When we apply a greedy forward selection scheme, the least-squares problem may be solved by an order-recursive update of the pseudoinverse in each iteration step. The computational time is linear with respect to the number of the selected training samples.

The spin- Heisenberg star with frustration: II. The influence of the embedding medium

We investigate the spin 1/2 Heisenberg star introduced in J. Richter and A. Voigt, J. Phys. A: Math. Gen. {\bf 27}, 1139 (1994). The model is defined by

A Hybrid HMM-Based Speech Recognizer Using Kernel-Based Discriminants as Acoustic Models

H=J_1 \sum_{i=1}^{N}{{\bf s}_0{\bf s}_i} + J_2 H_{R}\{{\bf s}_i\}

Sparse Kernel Logistic Regression using Incremental Feature Selection for Text-Independent Speaker Identification

;

Quantum phase transitions in spin systems

J_1,J_2 \ge 0

Speaker identification and verification using support vector machines and sparse kernel logistic regression

,

Limited Training Data Robust Speech Recognition Using Kernel-Based Acoustic Models

i=1,...,N

Direct calculation of the spin stiffness of the spin-12 Heisenberg antiferromagnet on square, triangular, and cubic lattices using the coupled-cluster method

. In extension to the Ref. we consider a more general

Integrable models of interacting quantum spins with competing interactions

H_{R}\{{\bf s}_i\}