{"id":9847,"date":"2024-03-27T10:13:36","date_gmt":"2024-03-27T10:13:36","guid":{"rendered":"\/\/www.8xbethub.com\/?p=9847"},"modified":"2024-03-27T10:14:13","modified_gmt":"2024-03-27T10:14:13","slug":"industry-news-humanoid-robot-converses-with-openai-language-ai","status":"publish","type":"post","link":"\/\/www.8xbethub.com\/2024\/03\/27\/industry-news-humanoid-robot-converses-with-openai-language-ai\/?lang=en","title":{"rendered":"Humanoid robot converses with OpenAI language AI"},"content":{"rendered":"\n

The unicorn tech company’s Figure AI robot is equipped with OpenAI’s image language model for natural human interaction.<\/p>\n\n\n\n

In a video released by Figure AI on March 13th, the company’s humanoid robot, Figure 01, can seamlessly converse with humans. The company states that the robot integrates the latest Visual Language Model (VLM) developed by OpenAI, which enables speech-to-speech interaction for communication.<\/p>\n\n\n\n

Specifically, the VLM utilizes OpenAI’s multimodal model with the ability to understand images, text captured from the robot’s camera, based on real-time voice conversation. These data are analyzed by the robot to provide responsive feedback to the interlocutor. This differs from GPT-4, which focuses on text-based prompts.<\/p>\n\n\n

\n
\"\"
The dialogue between the robot in Figure 01 and humans.<\/figcaption><\/figure><\/div>\n\n\n

Additionally, the robot also employs “learned hand gestures.” The system seamlessly collects and controls the arms’ movements. “The image processing network operates at a frequency of 10 Hz and generates 24-DOF actions (wrist poses and finger joint angles) at a frequency of 200 Hz,” according to Figure AI.<\/p>\n\n\n\n

A demo video shows that Figure 01 can interact with humans quite naturally, although the responses are not yet sufficiently fast. Business Insider evaluates the humanoid robot equipped with the leading company’s AI like OpenAI as a “powerful combination,” but “it may evoke a mix of awe and fear.”<\/p>\n\n\n\n

Figure AI asserts that the content in the video is based on the system’s learning process and is not remotely controlled.<\/p>\n\n\n\n

“Without witnessing it firsthand, it’s challenging to verify Figure AI’s claims. Or the company may have gone through hundreds of iterations to achieve such results,” TechRadar comments. “But if it’s 100% true, it would be fantastic.”<\/p>\n\n\n\n

Figure AI is a newly established startup in the US in 2022 but has made a strong impression on investors, becoming a unicorn valued at $2.6 billion by unveiling and continuously improving the humanoid robot Figure 01. This robot is expected to be deployed in manufacturing, transportation, logistics, warehousing, and retail – areas facing labor shortages.<\/p>\n\n\n\n

Source: vnexpress.net<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"The unicorn tech company’s Figure AI robot is equipped with OpenAI’s image language model for natural human interaction. In a video released by Figure AI on March 13th, the company’s humanoid robot, Figure 01, can seamlessly converse with humans. The company states that the robot integrates the latest Visual Language Model (VLM) developed by OpenAI,… View Article<\/a>","protected":false},"author":3,"featured_media":9845,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1,1],"tags":[254,362],"class_list":{"0":"post-9847","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","6":"hentry","7":"category-tin-nganh","9":"tag-mtavietnam-2-2","10":"tag-mtavietnam2024-2-2"},"acf":{"ngay_dang":"","add_tag_small":"","must_read":"khong","img_thumbnail_post_home":false},"yoast_head":"\n\n\n\n\n\n\n\n\n\n\n\n\t\n\t\n\n\n\t\n