How to Make a Burning Bridges AI Video

This walkthrough covers the whole Burning Bridges AI video workflow, from picking a photo to downloading the finished clip. There is no prompt to write and no timeline to edit. The scene is prepared, so the only decisions that change the result are the upload, the output ratio and the resolution.

Before you start

You need one photo and enough credits for the resolution you want. A photo with two people in it will not work, because the template has a single performer to replace.

Keep these in mind when you choose the photo:

• One person per photo, facing the camera.
• Even lighting on the face, with no hard shadow across it.
• No sunglasses, hats, or anything covering the face.
• A recent photo, so the identity reads clearly in the render.

You do not need to pose like the reference. The lip-sync posture, the hand movements and the walk through the room all come from the template.

Step 1: Upload one photo

In the generator, add a single portrait. The slot takes one image, up to 8 MB, and the uploader marks the file once it finishes. The counter next to the slot shows that one image is attached.

If you are testing an idea, a cheap 480P render is the fastest way to check whether the face reads well before you spend more credits.

Step 2: Pick the format and resolution

The format selector controls the output canvas:

• 4:5 matches the source clip and suits feed posts.
• 9:16 suits stories and vertical short-form.
• 16:9 suits YouTube and embedded posts.
• 1:1 suits mixed timelines and square posts.
• Auto follows the template frame when you do not want to choose.

The resolution selector controls quality and cost. 480P uses 50 credits, 720P Standard uses 100, and 1080P Pro uses 200. The required credits are shown in the form before you submit, so there is no surprise at checkout.

Step 3: Submit and review

Once the slot is filled, the generate button submits the paid task. The preview panel switches to a progress state while the provider works, and the finished clip appears in the stage as soon as it is ready. Recent renders stay in the strip underneath, so you can compare two photos side by side.

When a render is done, download it straight from the stage. The file keeps the template audio attached, so it is ready to post without re-timing.

What to do when a render fails

Provider failures are handled automatically. If a task fails, the unused credits are returned to your balance, and the tile in the history strip shows which task failed. The most common causes are a photo that the moderation layer flags and settings the provider does not support; both are listed in the error message under the preview.

Common questions

Can I use a character instead of a person? Yes. The same slot accepts a character portrait, which is how the Shrek and Thanos versions of the trend are made. The steps are identical.

Does it keep the audio? Yes. The template audio stays attached during generation, so the hook and the room noise land on the same beat as the source clip.

Is there a free tier? There is no free generation tier. The cheapest way in is a one-time pack from $9.90, and monthly subscriptions start at $19.