We are very interested in frustration šŸ˜” so we are building multi-modal models that combine both face and voice to gā€¦ https://t.co/ytV9x6FtMC