A unified speaker-dependent speech separation and enhancement system based on deep neural networks.
Tian GaoJun DuLi XuCong LiuLi-Rong DaiChin-Hui LeePublished in: ChinaSIP (2015)
Keyphrases
- speaker dependent
- neural network
- speech recognition
- phoneme recognition
- speaker identification
- speaker independent
- pattern recognition
- hidden markov models
- speech signal
- artificial neural networks
- automatic speech recognition
- gaussian mixture model
- noisy environments
- speech synthesis
- image processing
- language model
- speech recognizer
- speech recognition systems
- feature extraction
- speaker recognition
- digit recognition
- information retrieval
- visual information
- dialogue system
- machine learning