Why Doesn't My Spokesperson Photo Speak After I Add It to a HeyGen Studio Scene?

Determine Whether the Photograph Must Speak

A merchant might use the same approved spokesperson photograph in two advertisements. One displays the person's portrait beside a written statement. The other requires the person to deliver spoken promotional lines.

Those tasks have different production requirements. The first needs a static visual element; the second needs an avatar capable of participating in the speaking-video workflow.

HeyGen Studio distinguishes ordinary Assets from Photo Avatars. Uploading a photograph to Assets and placing it on the canvas does not produce a lip-synced avatar.

Decision branch: If the portrait should remain still, treat it as an image Asset. If the person must visibly speak, evaluate Photo Avatar creation through the Avatar section rather than repeatedly repositioning the static image.

Terminology note: HeyGen’s separate media-upload guide calls these ordinary items Media, formerly Assets. An older Studio instruction may therefore use Assets while the current upload panel says Media. The relevant distinction is still ordinary scene media versus an Avatar, not the label alone.

Separate Canvas Placement From Avatar Creation

A photograph appearing correctly on the canvas establishes that it is available as a visual element. It does not establish that the person has been configured as a speaking character.

When troubleshooting, identify the source role before changing scripts or audio. If the image was introduced through Assets, the missing lip movement may reflect the chosen object type rather than a problem with the written dialogue.

The documented Photo Avatar route is through the Avatar section. That workflow distinction should be recorded in the production brief so the next editor understands whether the photograph is a static graphic or an intended speaking subject.

A No Avatar in Video indication can also be intentional when the project is designed without a speaking avatar. Do not automatically treat every avatar-free scene as an error.

Illustrative Exercise: Two Testimonial Formats

Illustrative exercise — fictional merchant briefs, not executed HeyGen Studio sessions, avatar creations, or video tests.

Input A is a static testimonial card showing an appropriately authorized portrait, a quotation, and product imagery. No facial movement is required. The portrait can serve as an ordinary visual Asset.

Input B is a spokesperson advertisement in which an appropriately authorized person must visibly deliver a short script. The production team should evaluate the Photo Avatar workflow instead of assuming the static Asset will animate.

Decision branch: For A, approve the static composition. For B, confirm the speaking-avatar requirement and select the appropriate creation route before reviewing the finished performance.

Build a Clear Speaking-Role Handoff

Before editing, label every person image by its intended role: static portrait, background visual, or speaking spokesperson. Record whether the scene actually requires an avatar.

For static testimonial material, review portrait placement, quotation accuracy, and whether the presentation could mislead viewers about the speaker's participation.

For speaking content, separately review the intended script, authorized image and voice materials, and whether the resulting video communicates the approved message.

Do not present a generated actor as an actual customer or imply that avatar creation establishes testimonial authenticity. Obtain appropriate authorization for the materials used without assuming the software independently verifies permissions.

This is a source-role and workflow decision, not a generic preview-versus-export problem. The documented distinction does not guarantee a particular lip-sync result.

Illustrative Exercise — Static Portrait Versus Speaking Avatar

These are illustrative exercise rows. Save a copy and record your own observations; the example is not an executed acceptance test.

Download review worksheet (CSV)