Models used for the paper:
E. Moliner, F. Elvander and V. Välimäki,, "Zero-Shot Blind Audio Bandwidth Extension", submitted to IEEE Transactions on Audio, Speech, and Language Processing, 2023
They consists on unconditional diffusion models trained on
- MAESTRO (solo piano music)
- COCOChorales (strings)
- COCOChorales (woodwind)
- COCOChorales (brass)
Unable to determine this model's library. Check the
docs
.