Seven months after the discharge of ‘GPT-4o’, OpenAI added the real-time video understanding mode that was shown within the demo. Moreover, Santa’s voice was added to ChatGPT as a Christmas edition.
On the twelfth (local time), on the sixth day of Shipmas, OpenAI announced that it could add a real-time video understanding function to ChatGPT Advanced Voice Mode (AVM) through streaming.
When GPT-4o was released last May, this function understands and responds when an object is illuminated with a cellphone camera. For instance, they might explain various settings menus or solve math problems.
It was chosen as a key feature of the AI voice assistant, but its official release was postponed resulting from the interior development schedule being delayed. ‘Project Astra’, a synthetic intelligence (AI) agent introduced by Google the day past, also placed real-time video understanding at its core.
The brand new Vision AI feature might be began by tapping the voice icon next to the chat bar after which tapping the video icon in the underside left. Screen sharing can also be possible.
This feature starts today for ChatGPT Plus users, and the update is scheduled to be accomplished next week. It has been confirmed that the function has not yet been added in Korea.
Moreover, ChatGPT Enterprise and Edu are scheduled to be updated in January next yr, and is not going to be available in some countries reminiscent of the European Union (EU) and Switzerland in the interim.
As well as, OpenAI added Santa’s voice to AVM. Users can use ChatGPT by tapping the snowflake icon next to the chat bar.
This feature may even be launched from at the present time and will probably be in service until early January after which terminated. Each conversation you have got with Santa doesn’t affect your chat history or memory. It has been confirmed that it will not be currently applied in Korea.
Reporter Lim Da-jun ydj@aitimes.com
