Exploiting Cross-domain And Cross-Lingual Ultrasound Tongue Imaging Features For Elderly And Dysarthric Speech Recognition.
Shujie HuXurong XieMengzhe GengMingyu CuiJiajun DengTianzi WangXunying LiuHelen MengPublished in: CoRR (2022)
Keyphrases
- speech recognition
- cross domain
- cross lingual
- sentiment classification
- hidden markov models
- language model
- transfer learning
- automatic speech recognition
- language modeling
- speech synthesis
- machine translation
- feature set
- cross language
- knowledge transfer
- speech signal
- computer vision
- language independent
- machine learning
- pattern recognition
- text classification
- prior knowledge
- pairwise
- speaker identification
- information retrieval