4. create a transcriber use the speech-recognition pipeline and the model of your choice

→ View original post on X — @reach_vb