A Cross-Task Transfer Learning Approach to Adapting Deep Speech Enhancement Models to Unseen Background Noise Using Paired Senone Classifiers.
Sicheng WangWei LiSabato Marco SiniscalchiChin-Hui LeePublished in: ICASSP (2020)
Keyphrases
- background noise
- transfer learning
- speech enhancement
- machine learning algorithms
- noisy environments
- low contrast
- training set
- reinforcement learning
- speech signal
- labeled data
- semi supervised learning
- text mining
- text classification
- machine learning
- training samples
- noise reduction
- collaborative filtering
- target domain
- information extraction
- active learning
- adaptive filtering
- pairwise