Strategies in Transfer Learning for Low-Resource Speech Synthesis: Phone Mapping, Features Input, and Source Language Selection.
Phat DoMatt ColerJelske DijkstraEsther KlabbersPublished in: CoRR (2023)
Keyphrases
- transfer learning
- speech synthesis
- cross lingual
- cross domain
- labeled data
- knowledge transfer
- speech recognition
- feature extraction
- machine translation
- feature space
- text to speech
- text classification
- machine learning algorithms
- feature vectors
- information retrieval
- data mining
- source language
- target language
- domain adaptation
- machine learning
- training data
- class labels
- text categorization
- neural network
- text mining
- collaborative filtering
- knowledge discovery
- hidden markov models
- prior knowledge