Damage Control During Domain Adaptation for Transducer Based Automatic Speech Recognition.
Somshubra MajumdarShantanu AcharyaVitaly LavrukhinBoris GinsburgPublished in: SLT (2022)
Keyphrases
- domain adaptation
- automatic speech recognition
- speech recognition
- speech signal
- hidden markov models
- cross domain
- conversational speech
- semi supervised
- labeled data
- multiple sources
- transfer learning
- semi supervised learning
- sentiment classification
- target domain
- broadcast news
- co training
- document classification
- feature representation
- test data
- training samples
- maximum likelihood
- supervised learning
- probabilistic model