is it possible to add a PDF file to the context before or at the beginning of a live multimodal session?
I tried sending the file uri, but I may be mixing up api versions because I can’t get it to recognize the pdf file contents
Hi @chao_Lam , Welcome to the forum.
Currently, with the Live API, the limitation is that the model can process only text, audio, and video input.
Thanks. So, the file upload apis won’t work in conjunction with Live API, even before Live API session starts?
Also, is this just a current limitation, with plans to incorporate native PDF processing later?