Transfer Learning from Pre-trained Language Models Improves End-to-End Speech Summarization.
Kohei MatsuuraTakanori AshiharaTakafumi MoriyaTomohiro TanakaTakatomo KanoAtsunori OgawaMarc DelcroixPublished in: INTERSPEECH (2023)
Keyphrases
- end to end
- transfer learning
- language model
- pre trained
- speech recognition
- language modeling
- training data
- labeled data
- spoken term detection
- n gram
- training examples
- speech signal
- probabilistic model
- active learning
- reinforcement learning
- information retrieval
- cross domain
- semi supervised learning
- machine learning
- query expansion
- text classification
- automatic speech recognition
- learning algorithm
- control signals
- context sensitive
- text categorization
- text mining
- collaborative filtering
- supervised learning
- unlabeled data
- cross lingual
- real world
- computer vision
- broadcast news
- expectation maximization
- data mining