Spatial Audio Processing and Sound-Field Modeling

Sparse acoustic observations, neural and physics-informed models, generative methods, and sound-field and HRTF interpolation

We investigate spatial audio processing from limited acoustic observations and the modeling of spatial hearing. Research includes neural interpolation of head-related transfer functions across source directions and sparse sound-field estimation through distinct physics-informed and flow-matching approaches. A separate study proposes a virtual microphone for reproducing listening conditions at positions between two talkers and examines the influence of acoustic sources and visual information.

Joint work with Prof. Shoichi Koyama (NII) and collaborators

Demo

Related Publications

Ege Erdem, Shoichi Koyama, Tomohiko Nakamura, Orchisama Das, and Zoran Cvetkovic, “SF-Flow: Sound field magnitude estimation via flow matching guided by sparse measurements,” in Proceedings of International Workshop on Acoustic Signal Enhancement, Sept. 2026.
Karl Schrader, Shoichi Koyama, Tomohiko Nakamura, and Mirco Pezzoli, “Phase-retrieval-based physics-informed neural networks for acoustic magnitude field reconstruction,” in Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing, May 2026, pp. 15162–15166.
Ryan Niu, Shoichi Koyama, and Tomohiko Nakamura, “Head-related transfer function individualization using anthropometric features and spatially independent latent representations,” in Proceedings of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, Oct. 2025.
Yuki Ito, Tomohiko Nakamura, Shoichi Koyama, Shuichi Sakamoto, and Hiroshi Saruwatari, “Spatial upsampling of head-related transfer function using neural network conditioned on source position and frequency,” IEEE Open Journal of Signal Processing, vol. 6, pp. 1109–1123, Sept. 2025.
Shoichi Koyama, Juliano Ribeiro, Tomohiko Nakamura, Natsuki Ueno, and Mirco Pezzoli, “Physics-informed machine learning for sound field estimation,” Special Issue on Model-Based and Data-Driven Audio Signal Processing, IEEE Signal Processing Magazine, vol. 41, pp. 60–71, Nov. 2024.
Yuki Ito, Tomohiko Nakamura, Shoichi Koyama, and Hiroshi Saruwatari, “Head-related transfer function interpolation from spatially sparse measurements using autoencoder with source position conditioning,” in Proceedings of International Workshop on Acoustic Signal Enhancement, Sept. 2022.
Kazuhide Shigemi, Shoichi Koyama, Tomohiko Nakamura, and Hiroshi Saruwatari, “Physics-informed convolutional neural network with bicubic spline interpolation for sound field estimation,” in Proceedings of International Workshop on Acoustic Signal Enhancement, Sept. 2022.