Shuai Yu, Chenxing Li, Feng Deng, Xiaorui Wang: Rethinking Singing Voice Separation With Spectral- Temporal Transformer. APSIPA ASC 2021: 884-889