Deep Learning of Articulatory-Based Representations and Applications for Improving Dysarthric Speech Recognition.
Feifei XiongJon BarkerHeidi ChristensenPublished in: ITG Symposium on Speech Communication (2018)
Keyphrases
- speech recognition
- deep learning
- unsupervised feature learning
- deep belief networks
- hidden markov models
- language model
- speech synthesis
- unsupervised learning
- pattern recognition
- speech recognizer
- speech recognition technology
- speech signal
- automatic speech recognition
- machine learning
- mental models
- speaker identification
- weakly supervised
- higher order
- vocal tract
- speech recognition systems
- image segmentation
- neural network
- signal processing
- object detection
- speaker independent
- speaker adaptation