Synthesia Expert Sessions vs a Sume product walkthrough clip

Expert Sessions is the third Synthesia Sessions type, announced on October 1, 2026 as coming soon, in which an avatar guides users through products and processes step by step. It is not available to buy yet, so there is no price or limit to compare. What you can do today is price the recorded version of the same job on Sume: a 30-second walkthrough with a product image is $5.82 on Standard, $7.74 on Plus and $17.40 on Max.

Those clips do not answer follow-up questions. They do show a product with a presenter, which is the part many walkthroughs actually need.

What Synthesia has said so far

The launch post describes Roleplay Sessions and Survey Sessions as available on October 1, 2026, and Expert Sessions as coming soon. It says the platform covers 140+ languages with 230+ avatars and is available 24/7. It gives no Expert Sessions price, no limits and no date beyond coming soon, so anything beyond that description would be a guess.

A product walkthrough on Sume

The Avatar 1.0 request takes a product_image (a public HTTPS URL) and an optional scene, as a prompt or a photo. The catalog rate with a product image is $0.194 a second on Standard, $0.258 on Plus and $0.58 on Max. Without a product image, the rates are $0.184, $0.245 and $0.55. The docs set the job window at 4 to 60 seconds.

Splitting a process into steps

A step-by-step guide maps to one clip per step. Each step is short, so each stays under the 60-second cap and can be replaced alone when the product UI changes. A five-step setup at 20 seconds a step is 100 seconds, $19.40 on Standard with the product image on every step. Ask for the same avatar handle in each so the presenter does not change between steps.

A single job accepts one avatar and one shared scene, so changes of location mean separate jobs. The multi-scene recipe covers what can change inside one video.

When to wait for Expert Sessions

Wait if users arrive with their own questions about your product and you cannot predict them. Use clips if the steps are known and fixed, because a recorded clip costs the same for the first viewer and the ten-thousandth. If you later add a live layer, the clips still serve as the first answer before the live session starts.