This research on AR/MR audio engine comprises of three key processing blocks: the sensing block, the real sound control block, and the virtual sound rendering block. The sensing block gathers data on the user’s acoustic environment and user/device responses using various sensors, then relays this information to the real sound control and virtual sound rendering blocks. The real sound control block modifies and transforms the existing acoustic environment to match the user's preferences, creating a tailored auditory experience, while the virtual sound rendering block generates audio signals to recreate desired virtual soundscapes.
Head-related transfer function (HRTF) is crucial for creating immersive listening experiences in VR and AR applications, especially in the Metaverse, where users interact with near-field (NF) virtual objects. Traditional static stop-and-go HRTF measurement methods are time-consuming and tedious. This paper proposes a continuous measurement system that captures NF HRTFs efficiently in 45 seconds, with experiments showing its consistency and robustness on both head and torso simulator (HATS) and human listeners.
With the strong growth of assistive and personal listening devices, natural sound ren- dering over headphones is becoming a necessity for prolonged listening in multime-dia and virtual reality applications. The aim of natural sound rendering is to naturally recreate the sound scenes with the spatial and timbral quality as natural as possible, so as to achieve a truly immersive listening experience. However, rendering natural sound over headphones encounters many challenges. This tutorial article presents signal processing techniques to tackle these challenges to assist human listening.
Augmented reality (AR) integrates virtual and real-world environments, gaining prominence with wearable devices. While AR enhances reality perception in various domains like education, gaming, and navigation, it has historically focused more on visual aspects than auditory components. Addressing this gap, this paper introduces a new headphone design with dual binaural microphones to enable natural listening in AR. It employs adaptive filtering techniques to blend virtual and physical sound sources effectively, enhancing user experience and perceptual realism, as validated through subjective and objective evaluations using individualized binaural room impulse responses (BRIRs).
The diversity of today's playback systems demands flexible, efficient, and immersive sound reproduction in digital media. The proposed ambient spectrum estimation (ASE) framework improves primary-ambient extraction (PAE) by addressing severe extraction errors in sound mixtures with strong ambient components. Experimental results show that ASE-based PAE significantly outperforms existing methods, offering both accurate and computationally efficient solutions.
Audio Beam
The parametric loudspeaker offers highly directional sound projection without large arrays and can be combined with conventional speakers for an immersive audio experience. Its deployment in public spaces allows private messaging, reduces noise pollution, and attracts attention without the need for headphones. Digital signal processing enhances aural quality, shapes, and steers beams, and overcomes limitations in nonlinear acoustics and ultrasonic transducer technology, paving the way for improved, controllable audio beaming and efficient sound-focusing devices.