Boosting Spatial Information for Deep Learning Based Multichannel Speaker-Independent Speech Separation In Reverberant Environments.
Ziye YangXiao-Lei ZhangPublished in: APSIPA (2019)
Keyphrases
- spatial information
- deep learning
- speaker independent
- speech recognition
- speech signal
- unsupervised feature learning
- speaker identification
- speaker dependent
- hidden markov models
- automatic speech recognition
- speech recognizer
- sound source
- unsupervised learning
- machine learning
- speech recognition systems
- noisy environments
- language model
- weakly supervised
- digit recognition
- spatial relations
- pattern recognition
- mental models
- non stationary
- dialogue system
- n gram
- data mining