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