Embody Me Co., Ltd.
“Expression camera Voice2Face” released that allows you to convert voice to your own appearance with AI and Zoom conference etc. while doing housework
EmbodyMe Co., Ltd. has released “expression camera Voice2Face” that converts voice into your own appearance with AI and allows you to have a Zoom meeting in any situation, such as while doing housework. Did. (URL: https://xpressioncamera.com/)
In addition, by providing a means of expressing emotions that cannot be achieved face-to-face, such as LINE stickers, we will create a completely new communication that goes beyond face-to-face.
Up to now, we have provided “expression camera” that can read facial expressions and body movements with a camera in video conferencing such as Zoom and live distribution such as Youtube, replace your appearance with AI and communicate in real time. I’ve been
By further improving and developing this, even without a camera, it is possible to generate images in real time using only voice as input, and with the click of a button, it is possible to generate a variety of movements that cannot be realized in person, resulting in richer emotions. We have developed a new technology “Voice2Face” that can express , and have significantly renewed it.
[Image 1: https://prtimes.jp/api/file.php?c_id=24788&t=animationGifImage&f=5d3acafba2bdde8285a6833f8fd216ab.gif&img_id=1666718177689]
[Video 2: https://prtimes.jp/api/movieim.php?url=www.youtube.com/watch?v=JirX8djlTU0]
[Image 2: https://prtimes.jp/api/file.php?c_id=24788&t=animationGifImage&f=6bf38ae7417f3d8b0e2e0da0f4e3f352.gif&img_id=1666718223461] Due to the corona crisis, video communication such as video chat has rapidly spread in many social activities such as work, medical treatment, education, and events.
However, on the other hand, everyone feels that it is “difficult to communicate compared to face-to-face”, which is an urgent issue. A big reason why the Metaverse is getting so much attention these days is because it uses VR devices to try to solve these problems as close as possible to a face-to-face experience.
However, most people still do not have VR devices, and the problem of fatigue caused by wearing VR devices for a long time has not been solved. .
“xpression camera” does not try to solve the “difficulty of
communication compared to face-to-face” by bringing it closer to the face-to-face experience as it is in the Metaverse, but the fundamental reason why face-to-face is superior in the first place. We will return to and solve it from a virtual perspective that is different from face-to-face.
The reason why face-to-face is superior in the first place is the problem of “Zoom fatigue” and the lack of non-verbal information and means of expressing emotions.
‘Zoom fatigue’ is a phenomenon that makes you feel mentally and physically tired during video chat. Research has shown that this is due to the tension caused by the need to match and the feeling that others are constantly watching you because your self-view is displayed.
Also, compared to face-to-face communication, it is difficult to convey non-verbal information such as facial expressions, gestures, and hand gestures, so it is difficult to achieve smooth communication. Casual communication, which is more valued, is lost online.
“xpression camera” can generate images in real time according to your facial expressions and body movements just by preparing a single image of your choice, such as a photo of yourself. You can communicate without worrying about how you are seen.
In addition, the new technology “Voice2Face” to be released this time will be able to generate your appearance in real time from only your voice without using a camera, so you will be able to communicate in a relaxed posture and free environment. can be reduced or eliminated. You will be able to video chat in any situation, such as taking care of your children, doing household chores, or jogging.
[Image 3: https://prtimes.jp/api/file.php?c_id=24788&t=animationGifImage&f=4e56ca495a3219b803964563be96c7b5.gif&img_id=1666718477207] In addition, with the click of a button, users will be able to express rich emotions through their own images and avatars on the screen, including deformed movements that cannot be achieved through face-to-face communication, such as LINE stamps.
By doing so, just as LINE stickers brought non-verbal emotional expressions to text communication, video communication will bring new emotional expressions that have never been seen before, creating a completely new form of communication that goes beyond face-to-face communication.
[Image 4: https://prtimes.jp/api/file.php?c_id=24788&t=animationGifImage&f=929246e132b80946d58b89f154544b5e.gif&img_id=1666718612461] Recently, technology for generating images from text, such as Stable Diffusion, has attracted a great deal of attention.
Sequoia, one of America’s leading venture capital firms, predicts that “Generative AI”, which creates content with AI, will bring about a major social change similar to that of smartphones, and the market size will grow to several hundred trillion yen in the future. I’m here.
https://www.sequoiacap.com/article/generative-ai-a-creative-new-world/ Combining a technology that generates images from text such as Stable Diffusion with “expression camera” makes it possible to easily generate images.
With only the technology to generate images from text, the market was limited to illustrations, comics, and art, but by using “expression camera”, a wide range of video markets such as movies, TV, animation, live distribution, and video chat can expand the possibilities. Our technology is at least 50 times faster than any other competitor in the world, and we have the advantage that only our technology works on common machines in real time for video chat, live streaming, games, etc.
We will further develop these advantages and lead the world with technology and culture in the major transformation of “Generative AI”.
[Image 5: https://prtimes.jp/api/file.php?c_id=24788&t=animationGifImage&f=851444e75f87f30f67dabbf3fc4d31f2.gif&img_id=1666719143078] * You can create videos by moving the facial expressions of images generated by Stable Diffusion in real time.
Details about this release: