Using Optimal Ratio Mask as Training Target for Supervised Speech Separation.
Shasha XiaHao LiXueliang ZhangPublished in: CoRR (2017)
Keyphrases
- supervised learning
- early stopping
- test set
- hearing impaired
- optimal solution
- training stage
- training set
- semi supervised
- speech recognition
- information retrieval
- supervised training
- training phase
- target tracking
- feature selection
- training samples
- online learning
- speech signal
- moving target
- worst case
- lower bound
- natural language
- video sequences
- endpoint detection
- machine learning