Semi-supervised acoustic model training for speech with code-switching.
Emre YilmazMitchell McLarenHenk van den HeuvelDavid A. van LeeuwenPublished in: Speech Commun. (2018)
Keyphrases
- semi supervised
- supervised learning
- hearing impaired
- fully labeled
- training process
- active learning
- labeled data
- semi supervised learning
- speech recognition
- training set
- unlabeled data
- unsupervised learning
- training examples
- label propagation
- automatic speech recognition
- multi view
- source code
- pairwise
- test set
- data sets
- speech signal
- co training
- labeled examples
- emotion recognition
- hidden markov models
- supervised methods
- speech synthesis
- recognition engine
- neural network