Add to ORKGBayesian approaches to speaker adaptation are popular in Automatic Speech Recognition (ASR) systems. In most kinds of Bayesian adaptation, there are parameters whose prior distributions are assumed to be multivariate normal. This paper presents a methodology, which can test the hypothesis of multivariate normality. When applied to Maximum A Posterior (MAP) adaptation, we found that the real prior distributions of the mean vectors are far from normal, which are always assumed in the MAP procedure. This result implies that better choice of the prior form may improve the adaptation result. 1
Models dealing directly with the raw acoustic speech signal are an alternative to conventional featu...
Summarization: Speaker adaptation is recognized as an essential part of today’s large-vocabulary aut...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...
As the use of found data increases, more systems are being built using adaptive training. Here trans...
Recent advances in automatic speech recognition are accomplished by designing a plug-in maximum a po...
One important issue in speech recognition is the ability to handle variations caused by unseen speak...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...
This paper shows the results achieved by the Maxi-mum A Posteriori (MAP) speaker adaptation method i...
We previously introduced a new Bayesian predictive classi-fication (BPC) approach to robust speech r...
Summary: We consider estimating a probability density p based on a random sample from this density b...
Stochastic n-gram language models have been successfully applied in continuous speech recognition fo...
The speaker-dependent HMM-based recognizers gives lower word error rates in comparison with the corr...
In this paper, a new method called Maximum Likelihood General Regression (MLGR) is introduced for sp...
Abstract—In this paper, a new hierarchical Bayesian speaker adaptation method called HMAP is propose...
Models dealing directly with the raw acoustic speech signal are an alternative to conventional featu...
Summarization: Speaker adaptation is recognized as an essential part of today’s large-vocabulary aut...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...
As the use of found data increases, more systems are being built using adaptive training. Here trans...
Recent advances in automatic speech recognition are accomplished by designing a plug-in maximum a po...
One important issue in speech recognition is the ability to handle variations caused by unseen speak...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...
This paper shows the results achieved by the Maxi-mum A Posteriori (MAP) speaker adaptation method i...
We previously introduced a new Bayesian predictive classi-fication (BPC) approach to robust speech r...
Summary: We consider estimating a probability density p based on a random sample from this density b...
Stochastic n-gram language models have been successfully applied in continuous speech recognition fo...
The speaker-dependent HMM-based recognizers gives lower word error rates in comparison with the corr...
In this paper, a new method called Maximum Likelihood General Regression (MLGR) is introduced for sp...
Abstract—In this paper, a new hierarchical Bayesian speaker adaptation method called HMAP is propose...
Models dealing directly with the raw acoustic speech signal are an alternative to conventional featu...
Summarization: Speaker adaptation is recognized as an essential part of today’s large-vocabulary aut...
Real-life speaker verification systems are often implemented using client model adaptation methods, ...