Generate speech from text using a cloned voice
Experiment with small super OCR models here.
text-to-3D & image-to-3D