Unsupervised Pre-training of Bidirectional Speech Encoders via Masked Reconstruction.
Weiran WangQingming TangKaren LivescuPublished in: CoRR (2020)
Keyphrases
- hearing impaired
- supervised learning
- speech recognition
- three dimensional
- semi supervised
- unsupervised learning
- test set
- training set
- supervised training
- training samples
- deep architectures
- training process
- neural network
- speech synthesis
- reconstruction method
- reconstruction process
- training algorithm
- learning algorithm
- video encoder
- training phase
- video compression
- training examples
- high resolution