no code implementations • 21 Mar 2022 • Zewang Zhang, Yibin Zheng, Xinhui Li, Li Lu
To improve the accuracy and naturalness of synthesized singing voice, we design several specifical modules and techniques: 1) A deep bi-directional LSTM-based duration model with multi-scale rhythm loss and post-processing step; 2) A Transformer-alike acoustic model with progressive pitch-weighted decoder loss; 3) a 24 kHz pitch-aware LPCNet neural vocoder to produce high-quality singing waveforms; 4) A novel data augmentation method with multi-singer pre-training for stronger robustness and naturalness.
1 code implementation • 18 Jul 2019 • Yibin Zheng, Xi Wang, Lei He, Shifeng Pan, Frank K. Soong, Zhengqi Wen, Jian-Hua Tao
Experimental results show our proposed methods especially the second one (bidirectional decoder regularization), leads a significantly improvement on both robustness and overall naturalness, as outperforming baseline (the revised version of Tacotron2) with a MOS gap of 0. 14 in a challenging test, and achieving close to human quality (4. 42 vs. 4. 49 in MOS) on general test.