Hedra
: How to use it, features, and the business problems it solves
Add bookmark
What is Hedra?
A multimodal AI from Hedra (US) that generates "talking / singing character videos" from a single portrait image and audio. Its current flagship models are Hedra Omnia (full-scene control including camera and full-body motion) and Hedra Avatar (long-form up to 5 minutes), and it has expanded into a platform that also generates images, video and audio and bundles third-party models. Free tier available; credit-based.
Business problems it solves
About "Hedra"
What is Hedra
Hedra is an AI that generates lip-synced "talking / singing character videos" from a single portrait image and audio. Provided by Hedra, Inc. (US), you give it a photo and audio (text-to-speech, an upload, or a voice clone) and it produces a person video with matching mouth movements and expressions. More recently it has expanded into a "platform for visual intelligence" that also generates images, video and audio and bundles third-party models. It has over 3 million users and raised a $32M Series A led by a16z.
The models have moved on. The Character-3 referenced in older descriptions is the previous generation (March 2025); the current flagships are Hedra Omnia (short-form, full-scene control including camera and full-body motion) and Hedra Avatar (long-form, up to 5-minute one-shot). Character-3 remains as a lightweight option, but there is no "Character-4" (as of August 2026).
Turn a photo and a voice into a talking video
The core of Hedra is that a single image and audio are enough to make that person speak (or sing) naturally in a video. The mouth, expressions and head movements sync to the audio. With the current Omnia you can direct not just the face but camera movement, background and full-body gestures via text, while Avatar handles long-form talking heads up to 5 minutes.
Features
01Character video
- Hedra Omnia — full-scene control (camera, background, full body), up to 8s
- Hedra Avatar — one-shot long-form up to 5 minutes
- Character-3 — lightweight option for fixed-camera avatars
02Images and audio too
- Creative Studio — also spans third-party models like Kling, Flux and Nano Banana
- Voice/music generation & voice cloning — supply a voice to drive the video
03Collaboration & realtime
- Multiplayer canvas — create together as a team
- Realtime Avatar & API — realtime conversational avatars, developer API
How to use
-
Prepare a portrait image
Provide one image of the person or character you want to speak.
-
Specify the audio
Provide text to read aloud, or uploaded / cloned audio.
-
Pick a model and generate
Choose Omnia (short, expressive) or Avatar (long-form) and generate.
-
Export
Export the finished video (paid plans are watermark-free and allow commercial use).
Pricing
Credit-based, with monthly plans (USD; no annual plans listed).
| Plan | Monthly | Credits/mo | Highlights |
|---|---|---|---|
| Free | $0 | 100 | Trial (watermarked) |
| Basic | $15 | 1,500 | Commercial use, slower generation |
| Creator (popular) | $30 | 5,400 | Fast generation, commercial use |
| Professional | $75 | 14,400 | Fastest generation, commercial use, team features |
| Enterprise | Custom | Custom | Dedicated support, SSO, private deployment |
Paid plans are watermark-free and allow commercial use. Monthly credits do not roll over (separately purchased credit packs do not expire). Prices and credits can change, so check the official pricing page for the latest (as of August 2026).
How Hedra differs from other avatar-video AIs
Avatar-video AIs split into those strong at templated corporate presenter videos and those that generate freely from any image and audio. Hedra stands out for letting you craft expressions, full body and camera direction from a single image and any audio.
| Aspect | Hedra | HeyGen / Synthesia | D-ID |
|---|---|---|---|
| Input | Any image + any audio | Studio avatars + templates | Mainly face photos |
| Strength | Character expression, full-body/camera direction | Corporate presentations & training | Talking photos |
| Output | Talking/singing person video (up to 5 min) | Narration videos | Face-centric video |
For corporate avatar videos see HeyGen or Synthesia; for photo-based talking heads D-ID; for general video generation Runway ML. When you want expressive character videos from a single image and a voice, Hedra is a strong choice.
This page is based on information published on the official site as of August 2026. Model generations and pricing may change; please check the official site for the latest.
