Most audio systems simulate immersion through psychoacoustic tricks — narrowing the sweet spot, manufacturing phantom images, ignoring physics. SoniForge takes a different approach: we build the wave front. Literally. Sound field synthesis reconstructs the acoustic wave field physically, not perceptually.
The Problem with "Spatial Audio"
Stereo and surround create phantom images only from one position. Move your head, and the spatial effect collapses. Atmos and 360RA are binaural filters applied to headphone playback — they don't work with speakers.
Most "spatial audio" is applied in post as rendering filters. They simulate perception but ignore physics. Real acoustic environments require synthesis at the signal level, not an effect layered on top.
True WFS systems exist — at research labs and venues like the Las Vegas Sphere. Fraunhofer IDMT and L-Acoustics serve high-end installations. Accessible WFS for smart home and automotive OEM remains largely unbuilt.
Many "AI audio" products route processing through cloud APIs. Latency, privacy, connectivity — these are fundamental constraints for real-time spatial rendering in vehicles and homes. Edge DSP is the only viable path.
The Technology
Based on the Huygens-Fresnel principle: every point on a wave front is a source of a secondary spherical wave. A dense array of individually driven speakers can synthesize any arbitrary sound field by controlling delay and amplitude at each driver. The result is physically correct, not simulated.
Position, content, gain per sound source — not a channel bus
Delay + amplitude + EQ per speaker, computed in real time
Every listener in the room hears consistent spatial audio
Sound field synthesis applies the same physics across different scales and deployment contexts. We build for each vertical's specific requirements from the same foundation.
Premium smart speakers and sound bars. Multi-room coordination with shared spatial scenes. Adaptive room correction for real acoustic environments — not anechoic labs.
Cabin audio, EV exterior sound synthesis, zone-based spatial audio for passengers. Software-defined audio enables post-sale premium sound purchases — recurring revenue for OEM partners.
Small and medium venue installs — conference rooms, live performance, museum installations. Dante and AES67 network integration. Configurable via browser-based planning software.
Why Now
"You cannot distinguish the synthesized source from a real wave field at frequencies below the upper frequency limit of the system."— EMPAC Wave Field Synthesis, 2016 — the core physics that makes this possible
Multi-channel mic and driver arrays that cost thousands in 2015 now retail for tens of dollars. The economics of dense WFS arrays now work for consumer products.
ARM Cortex-A-class processors run real-time WFS kernels at latencies below 5ms. No cloud. No dependency. Spatial audio works at the edge, as it always should have.
No engine noise means the cabin audio is the product. Software-defined audio enables premium upsells, custom sound profiles, and branded acoustic signatures — a new revenue model.
Apple built AirPods on spatial audio. Dolby Atmos is now a marketing claim. The real hardware breakthrough — actual WFS, not simulated — is still available to build.
Physical sound field synthesis is not a feature request. It is a different physics. We have spent 25 years working toward this — building the signal processing, the hardware, the acoustic calibration. SoniForge is the next step.