Spaces:
Sleeping
Sleeping
Nick Martins
Demo safety: curated examples only (no upload), 0.3s RIR + iter cap → bounded, no stall
41bf0d3 | """Differentiable Feedback Delay Networks with Learnable Delay Lines. | |
| A clean, open-source PyTorch implementation of: | |
| A. I. Mezza, R. Giampiccolo, E. De Sena, A. Bernardini, | |
| "Data-driven room acoustic modeling via differentiable feedback delay | |
| networks with learnable delay lines," | |
| EURASIP Journal on Audio, Speech, and Music Processing, 2024(1):51. | |
| arXiv:2404.00082 | |
| An FDN is the classic Schroeder/Jot artificial-reverberation structure: a bank | |
| of delay lines coupled by a feedback matrix. Here *every* parameter is made | |
| differentiable -- crucially the delay-line lengths themselves -- so the whole | |
| network can be fit to a measured Room Impulse Response (RIR) by backpropagation. | |
| """ | |
| from .fdn import DifferentiableFDN | |
| from .losses import edc_loss, soft_edp_loss, edc, soft_edp, EDPConfig | |
| from .train import fit_fdn, TrainConfig | |
| __all__ = [ | |
| "DifferentiableFDN", | |
| "edc_loss", | |
| "soft_edp_loss", | |
| "edc", | |
| "soft_edp", | |
| "EDPConfig", | |
| "fit_fdn", | |
| "TrainConfig", | |
| ] | |
| __version__ = "0.1.0" | |