Supervision-Guided Codebooks for Masked Prediction in Speech Pre-training.
Chengyi WangYiming WangYu WuSanyuan ChenJinyu LiShujie LiuFuru WeiPublished in: INTERSPEECH (2022)
Keyphrases
- hearing impaired
- prediction accuracy
- vector quantization
- training process
- linear prediction
- multi layer perceptron
- training phase
- prediction algorithm
- prediction error
- radial basis function network
- training algorithm
- speech recognition
- training samples
- active learning
- multiscale
- machine learning
- n gram
- neural network
- training examples
- audio visual
- speech signal
- supervised learning
- discriminative power
- automatic speech recognition
- training set
- pattern recognition
- speech synthesis