Add to ORKGThis paper applies the recently proposed Extended Maximum Likelihood Linear Transformation (EMLLT) model in a Speaker Adaptive Training (SAT) context on the Switchboard database. Adaptation is carried out with maximum likelihood estimation of linear transforms for the means, precisions (inverse covariances) and the feature-space under the EMLLT model. This paper shows the first experimental evidence that significant word-error-rate improvements can be achieved with the EMLLT model (in both VTL and VTL+SAT training contexts) over a state-of-the-art diagonal covariance model in a difficult large-vocabulary conversational speech recognition task. The improvements were of the order of 1 % absolute in multiple scenarios. 1
Abstract—We present a new modeling approach for speaker recognition that uses the maximum-likelihood...
Abstract — Speaker space based adaptation methods for auto-matic speech recognition have been shown ...
The amount of training data has a crucial effect on the accuracy of HMM based meeting recognition sy...
This paper applies the recently proposed SPAM models for acoustic modeling in a Speaker Adaptive Tra...
This paper examines techniques for speaker normalisation and adaptation that are applied in training...
Recently, kernel eigenvoices were revisited using kernel representations of distributions for rapid ...
The goal of this thesis is to find new and efficient features for speaker recognition. We are mostly...
Summarization: Speaker adaptation is recognized as an essential part of today’s large-vocabulary aut...
In the paper we present two techniques improving the recognition accuracy of multilayer perceptron n...
Speaker adaptive training (SAT) is a useful technique for building speech recognition systems on non...
Inter-speaker variation can be coped rather well in speech recognition by speaker adaptation techniq...
Adaptation using linear transforms is well known to significantly improve the performance of speech ...
The paper investigates the integration of Heteroscedastic Linear Discriminant Analysis (HLDA) into ...
Speaker verification suffers from significant performance degradation on emotional speech. We presen...
This paper addresses the use of discriminative training criteria for Speaker Adaptive Training (SAT)...
Abstract—We present a new modeling approach for speaker recognition that uses the maximum-likelihood...
Abstract — Speaker space based adaptation methods for auto-matic speech recognition have been shown ...
The amount of training data has a crucial effect on the accuracy of HMM based meeting recognition sy...
This paper applies the recently proposed SPAM models for acoustic modeling in a Speaker Adaptive Tra...
This paper examines techniques for speaker normalisation and adaptation that are applied in training...
Recently, kernel eigenvoices were revisited using kernel representations of distributions for rapid ...
The goal of this thesis is to find new and efficient features for speaker recognition. We are mostly...
Summarization: Speaker adaptation is recognized as an essential part of today’s large-vocabulary aut...
In the paper we present two techniques improving the recognition accuracy of multilayer perceptron n...
Speaker adaptive training (SAT) is a useful technique for building speech recognition systems on non...
Inter-speaker variation can be coped rather well in speech recognition by speaker adaptation techniq...
Adaptation using linear transforms is well known to significantly improve the performance of speech ...
The paper investigates the integration of Heteroscedastic Linear Discriminant Analysis (HLDA) into ...
Speaker verification suffers from significant performance degradation on emotional speech. We presen...
This paper addresses the use of discriminative training criteria for Speaker Adaptive Training (SAT)...
Abstract—We present a new modeling approach for speaker recognition that uses the maximum-likelihood...
Abstract — Speaker space based adaptation methods for auto-matic speech recognition have been shown ...
The amount of training data has a crucial effect on the accuracy of HMM based meeting recognition sy...