Utterance-level Permutation Invariant Training with Discriminative Learning for Single Channel Speech Separation.
Cunhang FanBin LiuJianhua TaoZhengqi WenJiangyan YiYe BaiPublished in: ISCSLP (2018)
Keyphrases
- highly efficient
- discriminative learning
- single channel
- speech recognition
- multi channel
- sound source
- speech signal
- structured prediction
- face recognition
- audio visual
- generative model
- pattern recognition
- prior information
- random variables
- training samples
- training set
- object recognition
- wiener filter
- computer vision