Suzumushi Updates SoundObject to v4.1 - Introduces Machine-Learning-Based Generalized HRTF
Suzumushi has updated SoundObject to v4.1 for Windows.
SoundObject generates binaural audio with a sense of three-dimensional sound localization from a monaural sound source and its positional information. It supports 3D spatial audio for both headphones and stereo speakers.
The Head-Related Transfer Function (HRTF) is a key technology for three-dimensional audio. However, several sound quality issues must be addressed for wider adoption. SoundObject version 4.1 introduces a machine-learning-based Generalized HRTF (GHRTF) that achieves music production–level quality.
SoundObject is available for free from github.com/suzumushi0/SoundObject_binary
More information about the machine learning method can be found at suzumushi0.hatenablog.com/entry/SOv1/ML_EN
