Exploiting Semi-Supervised Training Through a Dropout Regularization in End-to-End Speech Recognition.
Subhadeep DeyPetr MotlícekTrung BuiFranck DernoncourtPublished in: INTERSPEECH (2019)
Keyphrases
- speech recognition
- end to end
- supervised training
- hidden markov models
- conditional random fields
- unsupervised learning
- training phase
- language model
- recognition process
- pattern recognition
- supervised learning
- speech recognizer
- speech signal
- neural nets
- training algorithm
- dependency parsing
- data mining
- bayesian networks
- reinforcement learning
- semi supervised learning
- visual features