Scene-Aware Audio Rendering via Deep Acoustic Analysis
This work addresses the challenge of realistic audio rendering for virtual reality or gaming applications, but it is incremental as it builds on existing geometric sound propagation methods.
The paper tackles the problem of capturing and reproducing real-world room acoustics using commodity devices by developing a deep learning method to estimate acoustic material properties from recorded audio and approximate geometry, then using these for audio rendering; a user study shows perceptual similarity between recorded and rendered audio.
We present a new method to capture the acoustic characteristics of real-world rooms using commodity devices, and use the captured characteristics to generate similar sounding sources with virtual models. Given the captured audio and an approximate geometric model of a real-world room, we present a novel learning-based method to estimate its acoustic material properties. Our approach is based on deep neural networks that estimate the reverberation time and equalization of the room from recorded audio. These estimates are used to compute material properties related to room reverberation using a novel material optimization objective. We use the estimated acoustic material characteristics for audio rendering using interactive geometric sound propagation and highlight the performance on many real-world scenarios. We also perform a user study to evaluate the perceptual similarity between the recorded sounds and our rendered audio.