AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.
Hey all, sharing this one, super happy how it came out 
You can download at http://semilla.ai/software (http://semilla.ai/software)
SEMILLA.AI v2 is a divination neural audio synthesis instrument. Audio from your track is encoded into the latent field of a neural audio model and re-grown in real time as a new voice with two A.I. models at once, one per channel.
You steer the timbre by casting seeds: gravity wells that pull a swarm of green particles, whose positions in the field become the sound.
SEMILLA.AI is not another preset synthesizer—it is an oracle you consult. Hand the reading over to physics simulations and chaotic orbits—Circle, Lorenz, Chua, Halvorsen—and the constellation composes melodic and textured patterns that never exactly repeat.
Freeze States, morph between them with an LFO, and sculpt the return with tape Delay, shimmer Reverb, and latency-compensated Dry/Wet.
Features
> Real-time neural synthesis — incoming audio is encoded into a neural audio model's latent space and decoded back as new sound in real time, directly inside your DAW.
> Two voices at once — independent neural audio models process the left and right channels. Browse and load your own trained models.
> Seeds & particles — seeds act as gravity wells (Pull / Pole / Spin). Green particles represent latent dimensions, and their positions shape the sound you hear.
> Physics as performance — Gravity, Friction, Bounce, Turbulence, Speed, and Range define how the particle swarm moves through the field.
> Composing through orbits — Circle and strange attractors (Lorenz, Chua, Halvorsen...) guide the seeds. A per-seed X/Y pattern matrix generates melodic and textured structures automatically.
> Latent manifolds — selectable mapping modes reshape how seed movement travels through latent space, from smooth interpolation to exploratory jumps.
> Per-dimension control — adjust Bias and Scale for each latent dimension, with randomize, reset, and flip controls available per side.
> States — capture seed constellations, mark A/B points, and morph between them with a shaped LFO controlling rate and depth.
> FX chain — stereo width, tape delay (per-side time, tempo sync, ping-pong), shimmer reverb (Size / Decay / Damp / Diffusion), and a final latency-compensated Dry/Wet control.
> Presets — save and recall complete configurations: models, constellation, physics, and FX.
Format & Compatibility:
VST3 : Mac Silicon
VST3: Windows 10/11.
64-bit.
Pricing
Complete License — €49
7 AI models included
Extra License — €99
12 AI models included
SEMILLA.AI v2 is developed by SEMILLA.AI Studio, an audio developer studio working at the intersection of artificial intelligence and electronic music.
SEMILLA began as research into real-time neural audio synthesis and evolved into a practice of divination—reading sound through systems of chance, movement, and transformation. Version 2 explores chaotic systems: determinism without prediction.
