Speaker Adaptive Training and Mixup Regularization for Neural Network Acoustic Models in Automatic Speech Recognition.
Natalia A. TomashenkoYuri Y. KhokhlovYannick EstèvePublished in: INTERSPEECH (2018)
Keyphrases
- acoustic models
- automatic speech recognition
- speech recognition
- neural network
- broadcast news
- discriminative training
- hidden markov models
- speech signal
- pattern recognition
- speech recognizer
- training process
- noisy environments
- speaker independent
- language model
- speaker identification
- acoustic features
- feature extraction
- information retrieval
- feature space
- training data
- spontaneous speech