Self-supervised Rewiring of Pre-trained Speech Encoders: Towards Faster Fine-tuning with Less Labels in Speech Processing.
Hao YangJinming ZhaoGholamreza HaffariEhsan ShareghiPublished in: CoRR (2022)
Keyphrases
- speech processing
- fine tuning
- pre trained
- training data
- speech recognition
- training examples
- signal processing
- speaker identification
- natural language processing
- multimedia systems
- artificial intelligence
- english text
- small number
- machine learning
- noisy environments
- pairwise
- automatic speech recognition
- speech signal
- variable length
- supervised learning
- training set
- support vector
- information extraction
- control signals
- learning algorithm
- multimedia
- computer vision
- broadcast news
- active learning