Automatic Speech Recognition Model Adaptation to Medical Domain Using Untranscribed Audio.
Askars SalimbajevsJurgita Kapociute-DzikienePublished in: DB&IS (2022)
Keyphrases
- medical domain
- automatic speech recognition
- broadcast news
- speech recognition
- acoustic features
- speech signal
- domain specific
- speaker identification
- hidden markov models
- conversational speech
- spontaneous speech
- clinical practice
- speech retrieval
- word error rate
- text mining
- multimedia
- spoken document retrieval
- medical records
- clinical trials
- speech corpus
- noisy environments
- information extraction
- audio visual
- pattern recognition
- information systems
- metadata
- image segmentation
- speech synthesis
- feature extraction
- active learning
- general purpose