Vital JEPA V7

Stereo 22.05 kHz, six-second audio encoder initialized from V6 and trained on a mixture of native Vital interventions and synthetic one-shot articulations. V7 adds explicit articulation and pitch objectives while retaining acoustic-descriptor and exact-control intervention losses.

The selected checkpoint is step 16,000: it was chosen as the best balanced checkpoint across reconstruction, acoustic correlation, control identification/direction, retrieval, articulation, and pitch probes.

See evaluation/ for checkpoint comparison and the V6-to-V7 forward pass on guitar, bass-pluck, and snare test sounds.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support