Reuse one AI spokesperson across videos with an avatar handle
A Sume avatar_handle is a stable name for a ready avatar. Sume strips a leading @, so you create once and call the handle in every talking-video request.
To reuse one AI spokesperson on Sume, create the avatar once with an avatar_handle, then pass the same handle in every talking-video request. A handle may start with @; Sume normalizes it and stores it without the @. Creating the avatar is a one-time, job-based step, and videos reuse it.
Create once, render many times
The Models overview describes Avatar 1.0 as two steps, and the handle is the link between them.
| Step | Route | Notes |
|---|---|---|
| Create | POST /v1/avatar-1.0/generate | Top-level avatar_handle plus input of type prompt, props or photo |
| List or read | GET /v1/avatar-1.0/avatars and /avatars/:id | Same shape as the older /v1/avatars |
| Render | POST /v1/avatar-1.0/talking-video | Send avatar_handle with script or video_inputs |
Why handles beat ids
Docs recommend a stable handle so that your app or agent has a simple name to use again later and does not depend only on a generated id. In practice this means a content calendar can refer to product_host or studio_presenter rather than a stored identifier, and a teammate or an agent can pick the right face from a list.
If the spokesperson must be unique per campaign, create one handle per campaign. If you want one face everywhere, use one handle and vary only the script, the scene prompt and the optional product_image.
Limits worth knowing
- One final video resolves to one avatar; the current execution does not mix several avatars in a video.
- Multi-scene video_inputs are expected to share one scene background.
- Avatar creation is billed separately from videos, as a one-time step.
- The handle must belong to a ready avatar; poll the creation job until it completes.
- Compatibility aliases such as /v1/models/sume/avatar/v1.0/runs still work, but /v1/avatar-1.0/generate is preferred.
A naming habit
Use lowercase words and underscores, name by role (reference_presenter, product_host), and avoid putting a real person's name in a handle that other people can see. Check Create new avatar for the exact input shapes.
Governance for shared spokespeople
If several teammates or agents render videos, keep a single list of approved handles and what each is for. The list endpoint returns the avatars in your workspace, so use it as the source of truth instead of a spreadsheet that can drift.
Rotate by creating a new avatar rather than editing an old one, and retire handles you no longer want to use. Videos already rendered keep working, since they are separate artifacts.
Avoid claiming a synthetic spokesperson is a real employee or customer. Label the clip as AI where rules require, and review your platform's policy on synthetic media before you publish at scale.
Sources
Related posts
More in Sume Avatar 1.0
- Add a silent beat to an AI avatar video with voice type silence
In Sume multi-scene avatar videos a scene with voice type silence is a pause with no speech. It needs a duration and rejects script text. Rules and example.
- Sume Avatar 1.0 ratios vs TikTok ad ratios: three overlap, two do not
Avatar 1.0 makes 1:1, 3:4, 9:16, 4:3 or 16:9 at 720p, 4 to 60 s. TikTok's non-Spark ad page lists 9:16, 16:9 and 1:1, so 3:4 and 4:3 need a crop.
- Does Sume Avatar 1.0 speak Spanish or Korean? English only in code
The Avatar 1.0 talking-video prompt on main says English only. What that means for Spanish or Korean lines, and the audio-driven route to test instead.
- Synthesia needs a live consent clip: what do Sume photo avatars need?
Synthesia says personal avatars need a live consent recording. Sume's photo avatar takes an HTTPS image URL, so your own release process must cover consent.
Written by Sume