FMLLR

Feature space Maximum Likelihood Linear Regression (fMLLR) is a widely used technique for speaker adaptation in HMM-based speech recognition.
Wikipedia

Topic mentions per year

Topic mentions per year

2002-2017
0246820022017

Papers overview

Semantic Scholar uses AI to extract papers important to this topic.
2018
2018
When an automatic speech recognition ASR system is deployed for real-world applications, it often receives only one utterance at… (More)
  • figure 2
  • figure 1
  • table I
  • figure 3
  • figure 5
Is this relevant?
2016
2016
In this paper, we propose the use of deep neural networks (DNN) as a regression model to estimate feature-space maximum… (More)
  • figure 1
  • figure 2
  • figure 3
  • figure 4
  • table 1
Is this relevant?
2016
2016
We describe a collection of acoustic and language modeling techniques that lowered the word error rate of our English… (More)
  • figure 1
  • table 2
  • table 1
  • figure 2
  • table 3
Is this relevant?
2014
2014
We introduce recurrent neural networks (RNNs) for acoustic modeling which are unfolded in time for a fixed number of time steps… (More)
  • figure 1
  • figure 2
  • figure 3
  • figure 4
  • table 1
Is this relevant?
Highly Cited
2013
Highly Cited
2013
We propose to adapt deep neural network (DNN) acoustic models to a target speaker by supplying speaker identity vectors (i… (More)
  • figure 1
  • figure 2
  • table II
  • table I
Is this relevant?
2013
2013
Adaptation techniques are necessary in automatic speech recognizers to improve a recognition accuracy. Linear Transformation… (More)
  • figure 1
  • table 1
Is this relevant?
Highly Cited
2013
Highly Cited
2013
Deep Convolutional Neural Networks (CNNs) are more powerful than Deep Neural Networks (DNN), as they are able to better reduce… (More)
  • table 1
  • table 4
  • table 2
  • table 3
  • table 5
Is this relevant?
2012
2012
In this paper, we propose a novel representation of the FMLLR transform. This is different from the standard FMLLR in that the… (More)
  • table 1
  • table 2
Is this relevant?
2011
2011
This paper presents a memory efficient single pass speech recognizer that makes use of pre-computed FMLLR transformations for… (More)
  • table 1
  • table 2
  • table 3
Is this relevant?
2011
2011
This paper describes subspace constrained feature space maximum likelihood linear regression (FMLLR) for rapid adaptation. The… (More)
  • table 2
  • table 1
Is this relevant?