Neural Fourier Shift for Binaural Speech Rendering
This addresses the problem of domain generalization for binaural audio rendering, which is important for applications like virtual reality and audio processing, though it appears incremental as it builds on existing synthesis approaches.
The paper tackles binaural speech rendering from monaural audio by proposing Neural Fourier Shift (NFS), a network that operates in Fourier space to predict delays and scales of early reflections based on geometric time delays. Results show NFS performs comparably to previous methods on benchmark datasets while using 25 times less memory and 6 times fewer calculations.
We present a neural network for rendering binaural speech from given monaural audio, position, and orientation of the source. Most of the previous works have focused on synthesizing binaural speeches by conditioning the positions and orientations in the feature space of convolutional neural networks. These synthesis approaches are powerful in estimating the target binaural speeches even for in-the-wild data but are difficult to generalize for rendering the audio from out-of-distribution domains. To alleviate this, we propose Neural Fourier Shift (NFS), a novel network architecture that enables binaural speech rendering in the Fourier space. Specifically, utilizing a geometric time delay based on the distance between the source and the receiver, NFS is trained to predict the delays and scales of various early reflections. NFS is efficient in both memory and computational cost, is interpretable, and operates independently of the source domain by its design. Experimental results show that NFS performs comparable to the previous studies on the benchmark dataset, even with its 25 times lighter memory and 6 times fewer calculations.