OSSEM: one-shot speaker adaptive speech enhancement using meta learning.
Cheng YuSzu-Wei FuTsun-An HsiehYu TsaoMirco RavanelliPublished in: INTERSPEECH (2022)
Keyphrases
- meta learning
- speech enhancement
- inductive learning
- learning tasks
- model selection
- vocal tract
- machine learning algorithms
- single channel
- data mining
- noisy environments
- noise reduction
- decision trees
- feature selection
- signal to noise ratio
- base classifiers
- machine learning
- speech recognition
- metamodel
- multi channel
- speech synthesis
- reinforcement learning
- prior knowledge
- background noise
- learning algorithm