Omnihuman AI

OmniHuman 1.5 creates talking, singing, and storytelling videos from a single image and voice. It targets creators, studios, and brands that need film-grade digital humans with realistic lip-sync, emotion, and cinematic motion. The system supports humans, anime, and pets, and offers multi-character scenes. It emphasizes accessible controls and credit-based generation without a subscription.

The tool works by taking a photo and an audio track as input, then synthesizing lip movements, facial expressions, and camera motion to produce a finished video. Optional prompts refine actions, camera direction, and scene events. It runs across OmniHuman 1.5 and 2.0 with credit-based generation and a focus on ease of use for independent creators and teams.

OmniHuman 1.5 offers a set of core capabilities designed to deliver film-grade performances. It supports rhythmic and emotional performances, context-aware actions aligned to spoken intent, and text-guided control of actions and camera motion. It enables single- and multi-person scenes with diverse subject support and consistent motion across visual styles.

The pricing page describes one-time credit packs and how credits are used for each second of generated video. It notes that credits can be used across OmniHuman 1.5 and 2.0, that purchases are not subscriptions, and that VAT is excluded and calculated at checkout. It also mentions Stripe as the payment method and refund information, along with support contact channels.

Use cases

• Talking avatars, singing performances, cinematic acting, and multi-character scenes from a single photo and voice.