Subspace LHUC for Fast Adaptation of Deep Neural Network Acoustic Models.
Lahiru SamarakoonKhe Chai SimPublished in: INTERSPEECH (2016)
Keyphrases
- neural network
- acoustic models
- speech recognition
- pattern recognition
- principal component analysis
- hidden markov models
- dimensionality reduction
- low dimensional
- bayesian networks
- high dimensional
- training process
- broadcast news
- feature selection
- feature extraction
- supervised learning
- principal components
- deep learning
- data sets