WZRD turns an audio track into a music video. The site says it analyses the audio for percussion and harmonic elements and uses neural networks to generate matching visuals, which are then rendered and combined with the original track. A third-party walkthrough describes the workflow: pick an aesthetic style, generate a set of AI images, choose about a dozen, arrange them into sequences, preview a 30-second cut, then render the full length.
Tying the visuals to the audio is the sensible design. The usual failure of generated video is footage that has no relationship to the music; here beat and harmony analysis drives the output, and the selection step keeps the creator in control of which images make the cut. The stated audience is musicians, advertisers and event organisers, which fits short promotional pieces and live visuals better than long-form video. The workflow centres on generated images arranged in sequences, so expect stylised motion rather than filmed footage.
Pricing is the open question. The marketing page publishes none, and a hands-on review reported credits sold at $19.27 each with no clear way to estimate the credits a full render consumes, which the reviewer called a deal-breaker. The site carries a 2024 copyright and names no company or team, so treat WZRD as a small product: use the free account and the preview before buying credits, and get the per-render credit cost in writing.
