https://scholars.lib.ntu.edu.tw/handle/123456789/607161
標題: | Investigating on incorporating pretrained and learnable speaker representations for multi-speaker multi-style text-to-speech | 作者: | Chien C.-M Lin J.-H Huang C.-Y Hsu P.-C HUNG-YI LEE |
關鍵字: | Few-shot;Multi-speaker text-to-speech;Speaker representation;Clone cells;Embeddings;Generalization ability;Speaking styles;Text to speech;Voice conversion;Signal processing | 公開日期: | 2021 | 卷: | 2021-June | 起(迄)頁: | 8588-8592 | 來源出版物: | ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | 摘要: | The few-shot multi-speaker multi-style voice cloning task is to synthesize utterances with voice and speaking style similar to a reference speaker given only a few reference samples. In this work, we investigate different speaker representations and proposed to integrate pretrained and learnable speaker representations. Among different types of embeddings, the embedding pretrained by voice conversion achieves the best performance. The FastSpeech 2 model combined with both pretrained and learnable speaker representations shows great generalization ability on few-shot speakers and achieved 2nd place in the one-shot track of the ICASSP 2021 M2VoC challenge. ? 2021 IEEE |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85106397888&doi=10.1109%2fICASSP39728.2021.9413880&partnerID=40&md5=9637afe10b6f5e4be8b287e41c0f7113 https://scholars.lib.ntu.edu.tw/handle/123456789/607161 |
ISSN: | 15206149 | DOI: | 10.1109/ICASSP39728.2021.9413880 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。