Deep Learning-Based Acoustic Feature Representations for Dysarthric Speech Recognition.
M. LathaM. ShivakumarG. ManjulaM. HemakumarM. Keerthi KumarPublished in: SN Comput. Sci. (2023)
Keyphrases
- deep learning
- feature representations
- speech recognition
- speech recognition systems
- speaker independent
- acoustic models
- image representation
- feature representation
- deep belief networks
- generative model
- unsupervised learning
- hidden markov models
- language model
- speech recognizer
- speech signal
- pattern recognition
- automatic speech recognition
- low level features
- mental models
- weakly supervised
- machine learning
- image classification
- probabilistic model
- information retrieval
- feature extraction
- multiscale
- object recognition
- image retrieval
- low dimensional
- signal processing
- co occurrence
- decision trees
- semi supervised