Unsupervised Active Learning: Optimizing Labeling Cost-Effectiveness for Automatic Speech Recognition.
Zhisheng ZhengZiyang MaYu WangXie ChenPublished in: INTERSPEECH (2023)
Keyphrases
- automatic speech recognition
- cost effectiveness
- active learning
- cost effective
- semi supervised
- supervised learning
- speech recognition
- unsupervised learning
- speech signal
- hidden markov models
- word error rate
- conversational speech
- active learning strategies
- recognition errors
- broadcast news
- e learning
- speech retrieval
- sample selection
- training set
- spontaneous speech
- machine learning
- spoken words
- unlabeled data
- learning algorithm
- word recognition
- noisy environments
- labeled data
- active learning framework
- speech corpus
- low cost
- relevance feedback
- learning process
- acoustic features
- multiscale
- bayesian networks
- information retrieval