Multi-Task CTC Training with Auxiliary Feature Reconstruction for End-to-End Speech Recognition.
Gakuto KurataKartik AudhkhasiPublished in: INTERSPEECH (2019)
Keyphrases
- end to end
- speech recognition
- multi task
- multi task learning
- isolated word
- learning tasks
- hidden markov models
- acoustic models
- multi class
- language model
- pattern recognition
- speech recognizer
- speech signal
- supervised learning
- speech recognition systems
- feature selection
- transfer learning
- training set
- learning problems
- speaker independent
- training examples
- feature vectors
- feature set
- image processing
- pairwise