Search
Nearly Half of Participants Mistook Tavus's New AI Model Video Conversation for a Real Person
2 day ago /
Read about 0 minute
Author：小编
Tavus, an AI video company based in San Francisco, USA, has released a new model called Griffin-Lite. Tests have shown that nearly half of the participants mistakenly believed the person on the other end of a one-minute video call generated by this model was a real person, with an accuracy rate far surpassing the 2.4% of the previous generation system. Tavus claims that this is the first real-time video conversation model to pass the video Turing test, defined as a human interaction model. The model can respond in real-time with affirmations or interruptions when both parties speak simultaneously, mimicking the behavior of a real conversation.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
