E2E-DASR: End-to-end deep learning-based dysarthric automatic speech recognition.
Ahmad S. AlmadhorRizwana IrfanJiechao GaoNasir SaleemHafiz Tayyab RaufSeifedine Nimer KadryPublished in: Expert Syst. Appl. (2023)
Keyphrases
- end to end
- deep learning
- automatic speech recognition
- speech recognition
- unsupervised learning
- speech signal
- broadcast news
- hidden markov models
- unsupervised feature learning
- conversational speech
- machine learning
- weakly supervised
- speech retrieval
- congestion control
- pattern recognition
- mental models
- error rate
- speech sounds
- edge detection