Add notes
This commit is contained in:
21
README.md
Normal file
21
README.md
Normal file
@@ -0,0 +1,21 @@
|
||||
# A plain-language guide: multi-talk dialogue templates
|
||||
|
||||
What follows is a working checklist, written so you can follow it with any clear, front-facing image you are allowed to use.
|
||||
|
||||
Two-line conversations for a single photo with two front-facing characters (left and right). Used with [side-by-side talking clip](https://fameo.ai/lip-sync-generators/ai-multi-talk).
|
||||
|
||||
## Layout
|
||||
|
||||
One still, two faces, both looking forward, mouths visible. I do not crop one speaker off the edge.
|
||||
|
||||
## Templates
|
||||
|
||||
- A: "Did you bring the gift?" / B: "I brought the cake. The gift is your problem."
|
||||
- A: "Happy anniversary." / B: "We should eat first, then talk."
|
||||
- A: "Is this the feature?" / B: "That is the whole feature. We can go home."
|
||||
|
||||
Each turn should be a short sentence. Overlapping audio makes both mouths chatter.
|
||||
|
||||
Only use photographs of people who agreed to appear, or original characters. A staged conversation is not evidence of a real one. More context lives on [make a photo talk](https://fameo.ai).
|
||||
|
||||
If something looks off, a better-lit source image usually fixes more than a rewritten script.
|
||||
Reference in New Issue
Block a user