Multiple-target deep learning for LSTM-RNN based speech enhancement.
Lei SunJun DuLi-Rong DaiChin-Hui LeePublished in: HSCMA (2017)
Keyphrases
- deep learning
- recurrent neural networks
- speech enhancement
- multiple targets
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- data association
- speech signal
- visual tracking
- unsupervised learning
- neural network
- machine learning
- moving target
- particle filtering
- feed forward
- weakly supervised
- multiple objects
- particle filter
- speech recognition
- motion estimation
- mental models
- video sequences
- kalman filter
- viewpoint
- three dimensional