SC-CNN: Effective Speaker Conditioning Method for Zero-Shot Multi-Speaker Text-to-Speech Systems.
Hyungchan YoonChanghwan KimSeyun UmHyun-Wook YoonHong-Goo KangPublished in: IEEE Signal Process. Lett. (2023)
Keyphrases
- high accuracy
- high precision
- text to speech
- significant improvement
- probabilistic model
- prosodic features
- support vector machine
- computationally efficient
- detection method
- speaker recognition
- em algorithm
- preprocessing
- segmentation algorithm
- model selection
- speaker verification
- segmentation method
- mutual information
- object oriented
- classification accuracy
- pairwise