ImmersiveFlow: Stereo-to-7.1.4 spatial audio generation with flow matching

Immersive spatial audio has become increasingly critical for applications ranging from AR/VR to home entertainment and automotive sound systems. However, existing generative methods remain constrained to low-dimensional formats such as binaural audio and First-Order Ambisonics (FOA). Binaural rendering is inherently limited to headphone playback, while FOA suffers from spatial aliasing and insufficient resolution for high-frequency. To overcome these limitations, we introduce ImmersiveFlow, the first end-to-end generative framework that directly synthesizes discrete 7.1.4 format spatial audio from stereo input. ImmersiveFlow leverages Flow Matching to learn trajectories from stereo inputs to multichannel spatial features within a pretrained VAE latent space. At inference, the Flow Matching model predicted latent features are decoded by the VAE and converted into the final 7.1.4 waveform. Comprehensive objective and subjective evaluations demonstrate that our method produces perceptually rich sound fields and enhanced externalization, significantly outperforming traditional upmixing techniques. Code implementations and audio samples are provided at: https://github.com/violet-audio/ImmersiveFlow.

Sound field reconstruction using neural processes with dynamic kernels

Accurately representing the sound field with the high spatial resolution is critical for immersive and interactive sound field reproduction technology. To minimize experimental effort, data-driven methods have been proposed to estimate sound fields from a small number of discrete observations. In particular, kernel-based methods using Gaussian Processes (GPs) with a covariance function to model spatial correlations have been used for sound field reconstruction. However, these methods have limitations due to the fixed kernels having limited expressiveness, requiring manual identification of optimal kernels for different sound fields. In this work, we propose a new approach that parameterizes GPs using a deep neural network based on Neural Processes (NPs) to reconstruct the magnitude of the sound field. This method has the advantage of dynamically learning kernels from simulated data using an attention mechanism, allowing for greater flexibility and adaptability to the acoustic properties of the sound field. Numerical experiments demonstrate that our proposed approach outperforms current methods in reconstructing accuracy, providing a promising alternative for sound field reconstruction.