반응형
[Paper 리뷰] MaskCycleGAN-VC: Learning Non-Parallel Voice Conversion with Filling in Frames
MaskCycleGAN-VC: Learning Non-Parallel Voice Conversion with Filling in FramesNon-parallel voice conversion을 위한 Cycle-Consistent Adversarial Network-based 방식은 time-frequency structure를 capture 하는 능력이 부족함MaskCycleGAN-VCCycleGAN-VC2의 확장으로써 Filling in Frames를 통해 training 하여 얻어짐Filling in Frames를 사용하여 input mel-spectrogram에 temporal mask를 적용하고 converter가 surrounding frame을 기반으로 missing frame을 fillin..
Paper/Conversion
2024. 8. 22. 10:27
반응형