Creating a custom Viseme animation to get result of 2D SVG output like "redlips_front"

[azatech]

I'm looking to create custom viseme animations for lip syncing in my project.

Am I understanding the process correctly for creating an SVG with a custom 2D avatar?

To create my own viseme animation similar to 'redlips_front' - FE: 'bluelips_front', I'll likely need to follow these general steps:

1. Do I need to design the visemes - the specific lip shapes or visemes for the phonemes (all 22 visemes in our case)?

2. Do I need to design animations of visemes - create a sequence of frames or keyframes that animate the lips transitioning between the different visemes smoothly?

3. Next step is exporting the animation... Do I need to directly use my previously created animations with the Azure response.json? And Iterate one by one...
   json from actual Azure API:

[
  {
    "offset": 50.0,
    "id": 0
  },
  ....

3.2. I couldn't find where to submit (FE: to Azure API to pass/create new type: 'bluelips_front')? Is it possible?

P.S. In the end, I want to get an SVG animated output (just like with "redlips_front" now).
Please advise on how I can create an avatar (using visemes + design)

[pankopon]

@yulin-li @yinhew Please check and comment.

[azatech]

@yulin-li @yinhew
I've UPDATED the previous question.
Thank you in advance 👍

[yulin-li]

Hi @azatech thanks for your feedback.

The Azure TTS service doesn't support custom SVG for now. I will forward your feedback to our PM for future planning.