GenerTTS: Pronunciation Disentanglement for Timbre and Style Generalization in Cross-Lingual Text-to-Speech.
Yahuan CongHaoyu ZhangHaopeng LinShichao LiuChunfeng WangYi RenXiang YinZejun MaPublished in: CoRR (2023)
Keyphrases
- cross lingual
- text to speech
- machine translation
- cross lingual information retrieval
- language modeling
- language independent
- event extraction
- cross language
- text classification
- speech recognition
- parallel corpus
- language model
- news articles
- transfer learning
- english text
- mono lingual
- word processing
- parallel corpora
- machine translation system
- query translation
- document clustering
- information retrieval systems
- natural language processing
- learning algorithm
- data mining