{"id":23104,"date":"2024-11-15T00:16:23","date_gmt":"2024-11-14T16:16:23","guid":{"rendered":"https:\/\/www.1ai.net\/?p=23104"},"modified":"2024-11-14T21:17:52","modified_gmt":"2024-11-14T13:17:52","slug":"%e8%ae%af%e9%a3%9e%e6%98%9f%e7%81%ab%e5%a4%9a%e6%a8%a1%e6%80%81%e4%ba%a4%e4%ba%92%e5%a4%a7%e6%a8%a1%e5%9e%8b%e4%b8%8a%e7%ba%bf%ef%bc%8c%e6%95%b0%e5%ad%97%e4%ba%ba%e3%80%81%e8%af%ad%e9%9f%b3%e3%80%81","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/23104.html","title":{"rendered":"Cyberfire starfire multimodal interaction big model on line, digital people, voice, vision support one key call"},"content":{"rendered":"

\"Xunfei open platform\" public number this evening announced thatiFlytek Spark<\/a>Large models of multimodal interactions<\/a>Officially online, its realization expands from voice interaction toReal-time multimode interaction of audio and video streams<\/strong>The new \"multi-modal, super-anthropomorphic and personalized\" capability realizes three-in-one voice, visual and digital human interaction, and supports one-key invocation.<\/p>\n

\"a2bbc2f6j00smxzks001jd000nc00hip\"<\/p>\n

According to reports, Xunfei Starfire multimodal interaction large model debut of super anthropomorphic digital human technology, digital human torso and limb movementsAbility to accurately match voice content<\/strong>The AI \"comes to life\" by quickly generating expressions and actions. By unifying text, speech, and expressions, it is possible toSemantic consistency across modalities<\/strong>The modeling process is a very simple one, and thus allows for a realistic and coherent expression of the emotions of the larger model.<\/p>\n

It supports hyper-anthropomorphic and extremely fast interactions with theunified neural network<\/strong>Direct end-to-end modeling of speech-to-speech, faster and smoother response, keenly perceive changes in mood, also free to change according to instructionsRhythm, size and persona of the voice<\/strong>.<\/p>\n

It supports multi-modal visual interaction, can \"understand the world\", \"recognize everything\", more comprehensively perceive the specific background scene, logistics status and other information, a more accurate understanding of the task, and through voice, gestures, behavior, emotions, etc. to make a comprehensive judgment, make the appropriate response. appropriate response.<\/p>\n

As previously reported by IT Home, users can make voice and video calls with the digital person, who can realize a natural voice conversation with the user, and character expressions and other expressions can also match the spoken utterances. Starfire super anthropomorphic digital person also supports multimodal interaction.Allows digital people to recognize what's in the camera<\/strong>, such as the Monkey King and Ultraman standing together, the brand and function of face creams, and the category of flowers.<\/p>","protected":false},"excerpt":{"rendered":"

\"Xunfei Open Platform\" public number announced this evening that Xunfei Starfire multimodal interaction model is officially online, which is expanded from voice interaction to real-time multimodal interaction in audio and video streams, with new \"multimodal, super anthropomorphic and personalized\" capabilities, realizing three-in-one voice, visual, and digital human interaction, supporting one-click call. The new \"multimodal, super anthropomorphic and personalized\" capability realizes voice, visual and digital human interaction in one, and supports one-key call. According to reports, Xunfei Starfire multimodal interaction model debut super anthropomorphic digital human technology, digital human torso and limb movements can accurately match the voice content, and quickly generate expressions and movements, so that the AI \"lifelike\". By unifying text, voice and expression, cross-modal semantic consistency can be realized, thus making the emotional expression of the big model real and coherent. It supports super anthropomorphic high-speed interaction, and adopts unified neural network to directly realize the end-to-end modeling of speech to speech, with faster and smoother response, and it can be sensitive to<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[4935,362],"collection":[],"class_list":["post-23104","post","type-post","status-publish","format-standard","hentry","category-news","tag-4935","tag-362"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/23104","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=23104"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/23104\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=23104"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=23104"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=23104"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=23104"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}