Two-Stream Joint-Training for Speaker Independent Acoustic-to-Articulatory Inversion.
Jianrong WangJinyu LiuLi LiuXuewei LiMei YuJie GaoQiang FangPublished in: CoRR (2023)
Keyphrases
- speech recognition
- speaker independent
- acoustic models
- hidden markov models
- speaker dependent
- speech recognition systems
- speech recognizer
- speech signal
- language model
- multi stream
- automatic speech recognition
- pattern recognition
- speaker identification
- speech synthesis
- emotion classification
- training set
- discriminative training
- multi modal
- noisy environments
- speaker adaptation
- broadcast news
- training process
- bayesian networks
- acoustic features
- multimedia
- training data