CBLDNN-Based Speaker-Independent Speech Separation Via Generative Adversarial Training.
Chenxing LiLei ZhuShuang XuPeng GaoBo XuPublished in: ICASSP (2018)
Keyphrases
- speaker independent
- speech recognition
- acoustic models
- speaker dependent
- speech recognizer
- hidden markov models
- speaker identification
- discriminative training
- automatic speech recognition
- digit recognition
- speech recognition systems
- speech signal
- broadcast news
- language model
- speech synthesis
- spoken language
- speaker adaptation
- information retrieval
- emotion classification
- hearing impaired
- emotion recognition
- noisy environments
- generative model
- training set
- training process
- model selection
- maximum likelihood
- supervised learning
- low level
- training data
- machine learning