Marcel Katz, Martin Schafföner, Sven E. Krüger, Andreas Wendemuth
In this paper we investigate three approaches of calibrating and fusing output scores for speaker verification. Today’s speaker recognition systems often consist of several subsystems that use...
Combining Frame and Turn-Level Information for Robust Recognition of Emotions within Speech (2008)
Bogdan Vlasenko, Björn Schuller, Andreas Wendemuth, Gerhard Rigoll
Current approaches to the recognition of emotion within speech usually use statistic feature information obtained by application of functionals on turn- or chunk levels. Yet, it is well known that...
Using Support Vector Machines in a HMM based Speech Recognition System (2008)
Sven E. Krüger, Martin Schafföner, Marcel Katz, Edin Andelic, Andreas Wendemuth
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 nonoptimally the...
Bogdan Vlasenko, Björn Schuller, Andreas Wendemuth, Gerhard Rigoll
Abstract. Opposing the pre-dominant turn-wise statistics of acoustic Low-Level-Descriptors followed by static classification we re-investigate dynamic modeling directly on the frame-level in...
Abstract Dynamics of Temporal Difference Learning (2008)
In behavioural sciences, the problem that a sequence of stimuli is followed by a sequence of rewards r(t) is considered. The subject is to learn the full sequence of rewards from the stimuli, where...
Maximum Margin Classification on Convex Euclidean Metric Spaces (2008)
André Stuhlsatz, Hans-günter Meier, Andreas Wendemuth
Summary. In this paper, we present a new implementable learning algorithm for the general nonlinear binary classification problem. The suggested algorithm abides the maximum margin philosophy, and...
Marcel Katz, Sven E. Krüger, Martin Schafföner, Edin Andelic, Andreas Wendemuth
Abstract. In this paper we investigate two discriminative classification approaches for frame-based speaker identification and verification, namely Support Vector Machine (SVM) and Sparse Kernel...
Acoustic Modeling in the Philips Hub-4 Continuous-Speech Recognition System (2007)
Reinhold Haeb-Umbach, Xavier Aubert, Peter Beyerlein, Dietrich Klakow, Meinhard Ullrich, Andreas Wendemuth, ...
In this paper we describe some characteristics of the acoustic modeling used in the Philips continuous-speech recognition system for the DARPA Hub-4 1997 evaluation, which are related to robustness...
NEURAL NETWORKS REVISITED: A STATISTICAL VIEW ON OPTIMISATION AND GENERALISATION (2007)
statistical analysis Statistical methods can be applied to analysis of Neural Networks to come up with on-average results for robustness, capacity, and generalisation in the presence of certain...
Acoustic Modeling in the Philips Hub-4 Continuous-Speech Recognition System (2007)
Reinhold Haeb-umbach, Xavier Aubert, Peter Beyerlein, Dietrich Klakow, Meinhard Ullrich, Andreas Wendemuth, ...
In this paper we describe some characteristics of the acoustic modeling used in the Philips continuous-speech recognition system for the DARPA Hub-4 1997 evaluation, which are related to robustness...
Support vector machines as acoustic models in speech recognition (2007)
Sven E. Krüger, Martin Schafföner, Marcel Katz, Edin Andelic, Andreas Wendemuth
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...
Discriminative kernel classifiers in speaker recognition (2007)
Marcel Katz, Martin Schafföner, Edin Andelic, Sven E. Krüger, Andreas Wendemuth
The goal of automatic speaker recognition is to identify a speaker or to verify if a speaker is the person he claims to be. We present an overview of state-of-the-art speaker recognition systems...
Updates for nonlinear discriminants (2007)
Edin Andelić, Martin Schafföner, Marcel Katz, Sven E. Krüger, Andreas Wendemuth
A novel training algorithm for nonlinear discriminants for classification and regression in Reproducing Kernel Hilbert Spaces (RKHSs) is presented. It is shown how the overdetermined linear...
Martin Schafföner, Edin Andelic, Marcel Katz, Sven E. Krüger, Andreas Wendemuth
A novel training algorithm for sparse kernel density estimates by regression of the empirical cumulative density function (ECDF) is presented. It is shown how an overdetermined linear least-squares...
Updates for nonlinear discriminants (2007)
Edin Andelić, Martin Schafföner, Marcel Katz, Sven E. Krüger, Andreas Wendemuth
A novel training algorithm for nonlinear discriminants for classification and regression in Reproducing Kernel Hilbert Spaces (RKHSs) is presented. It is shown how the overdetermined linear...
Mathias Mamsch, Marcus Holmberg, Werner Hemmert, Advisor Prof, Dr. Andreas Wendemuth, ...
Automatic speech recognition on
Limited training data robust speech recognition using kernel-based acoustic models (2006)
Martin Schafföner, Sven E. Krüger, Edin Andelic, Marcel Katz, Andreas Wendemuth
Contemporary automatic speech recognition uses Hidden-Markov-Models (HMMs) to model the temporal structure of speech where one HMM is used for each phonetic unit. The states of the HMMs are...
Mixture of support vector machines for hmm based speech recognition (2006)
Sven E. Krüger, Martin Schafföner, Marcel Katz, Edin Andelic, Andreas Wendemuth
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...
A hybrid hmm-based speech recognizer using kernel-based discriminants as acoustic models (2006)
Edin Andelić, Martin Schafföner, Marcel Katz, Sven E. Krüger, Andreas Wendemuth
In this paper we propose a novel order-recursive training algorithm for kernel-based discriminants which is computationally efficient. We integrate this method in a hybrid HMM-based speech...
Marcel Katz, Martin Schafföner, Edin Andelic, Sven E. Krüger, Andreas Wendemuth
Logistic Regression is a well known classification method in the field of statistical learning. Recently, a kernelized version of logistic regression has become very popular, because it allows...
Kernel fisher discriminants as acoustic models in hmm-based speech recognition (2005)
Martin Schafföner, Edin Andelic, Marcel Katz, Sven E. Krüger, Andreas Wendemuth
While the temporal dynamic of speech can be handled very efficiently by Hidden Markov Models (HMMs), the classification of the single speech units (phonemes) is usually done with Gaussian probability...
Speech recognition with support vector machines in a hybrid system (2005)
Sven E. Krüger, Martin Schafföner, Marcel Katz, Edin Andelic, Andreas Wendemuth
While the temporal dynamics of speech can be represented very efficiently by Hidden Markov Models (HMMs), the classification of speech into single speech units (phonemes) is usually done with...
Modeling uncertainty of data observation (2001)
An approach is presented both theoretically and experimentally which overcomes a number of existing conceptual and performance problems in density estimation. The theoretical approach shows methods...
Automatic transcription of English broadcast news (1998)
Peter Beyerlein, Xavier Aubert, Reinhold Haeb-umbach, Dietrich Klakow, Meinhard Ullrich, Andreas Wendemuth, ...
In this paper the Philips Broadcast News transcription system is described. The Broadcast News task aims at the recognition of "found " speech in radio and television broadcasts...
LanguageModel Investigations Related to Broadcast News (1998)
Dietrich Klakow, Xavier Aubert, Peter Beyerlein, Reinhold Haeb-umbach, Meinhard Ullrich, Andreas Wendemuth, ...
In this paper we present some experiments that have been performed while developing language models for the PHILIPS Broadcast News system. Three main issues will be discussed: construction of...
Automatic transcription of English broadcast news (1998)
Peter Beyerlein, Xavier Aubert, Reinhold Haeb-umbach, Dietrich Klakow, Meinhard Ullrich, Andreas Wendemuth, ...
In this paper the Philips Broadcast News transcription system is described. The Broadcast News task aims at the recognition of \found " speech in radio and television broadcasts without any...
Language-Model Investigations Related To Broadcast News (1998)
Dietrich Klakow, Xavier Aubert, Peter Beyerlein, Reinhold Haeb-Umbach, Meinhard Ullrich, Andreas Wendemuth, ...
In this paper we present some experiments that have been performed while developing language models for the PHILIPS Broadcast News system. Three main issues will be discussed: construction of...
Stabilities in optimal cluster separation networks (1994)
Two clusters of normalized vectors are optimally separated in a neural network for which thresholds and weights are trained to give maximum pattern stability. The performance of local, iterative...
Optimisation in neural networks. (1994)
Thesis (Ph. D.)--University of Oxford, 1994.