AI + 3D audio
To realize a truly intelligent and adaptive audio rendering system, we aim to build a DNN architecture that can extract environmental information from audio signals and noises. Using the extracted information and original 3D audio data, we can regenerate the 3D audio scene best suits the given playback environment.
Sound of Things
Sound field manipulation
This line of study deals with the creation/reproduction/manipulation of sound/wave fields in various shapes for holographic auditory displays and synthesis of high-dimensional auditory impressions. The underlying principle is simple interference of multiple wave fronts produced by multiple transducers/loudspeakers, but for the generation of a sound field in a desired shape, multi-dimensional aspects of a sound field should be considered in both the analysis and synthesis stages. Array signal processing techniques for controlling various objective and subjective variables in real-time constitute the core of this research field.