Software & Data Downloads — S2RNF
Sim-to-Real Neural Field for HRTF personalization by converting HRTFs simulated from an individual’s 3D head mesh into realistic counterparts.
Python code for training and evaluating on the SONICOM dataset the sim-to-real neural field (S2RNF) proposed in our Interspeech 2026 paper.
-
MERL Contacts
-
Related Publications
- , "HRTF Personalization via Sim-to-Real Neural Field", Interspeech, DOI: 10.21437/Interspeech.2026-1391, September 2026, pp. 696-701.
BibTeX TR2026-135 PDF Software- @inproceedings{Masuyama2026sep2,
- author = {Masuyama, Yoshiki and Wichern, Gordon and Boeddeker, Christoph and Richter, Julius and Edo, Takahiro and Bhosale, Swapnil and {Le Roux}, Jonathan},
- title = {{HRTF Personalization via Sim-to-Real Neural Field}},
- booktitle = {Interspeech},
- year = 2026,
- pages = {696--701},
- month = sep,
- doi = {10.21437/Interspeech.2026-1391},
- issn = {2958-1796},
- url = {https://www.merl.com/publications/TR2026-135}
- }
- , "HRTF Personalization via Sim-to-Real Neural Field", Interspeech, DOI: 10.21437/Interspeech.2026-1391, September 2026, pp. 696-701.
Software & Data Downloads
Access software at https://github.com/merlresearch/s2rnf.