Hour One yes describes it as a "video processing company" that wants to replace cameras with code. The latest creation of him is the possibility for anyone to create a fully digital clone of him that speak on video without.
The company launched its digital clone technology in partnership with the YouTuber Taryn Southern . In the video you can see a completely digital creation, born from a proprietary AI-based process.
Video? Let's get the digital clone to do it
Hour One provides that in the next 5-7 years, 90% of the content will be synthetic. Generated using computers instead of cameras.
When the pandemic hit, production around the world stopped. People were looking for alternative ways to create content and I was curious about what could be produced with AI-generated videos. Experimenting with the digital clone makes us reflect on the future of identity and trust, and will undoubtedly change the future of production.Taryn Southern
To create her AI clone, Southern had to walk into a studio and stand in front of a green screen in order to be shot from multiple angles. He also uttered several sentences so that the program could replicate his voice. Few things: he read some lyrics and sang a song. The whole process took just seven minutes. Here is a backstage video.
From that process it was possible to create hundreds of videos in minutes, simply by sending text to the platform. No cameras, no shots to record, no audio to edit.
The advantages of the digital clone
Hour One argues that the benefits of having a digital clone outweigh the possible drawbacks. The company says that with this technology, content creators will see a drastic reduction in video production time and costs. Taryn jokes that she can now create new YouTube videos "without the real Taryn having to shower or leave her bed." And she is not wrong.
But there's more: any AI clone can speak multiple languages, easily distributing its content to multiple people around the world.
What's the difference with a deepfake?
It is important to distinguish this digital clone technology from a "deepfake". Deepfake takes a face and overlays it on existing or newly recorded footage. What Hour One does is create completely original content as if it were spoken by a real person. Hour One defines the result as a "photorealistic digital human".
While this process may not lend itself to all types of content (like comedy, for example, which relies heavily on performance and timing), Hour One argues that it could be very effective for news formats, for which the attention is paid to the timeliness and quality of writing and reporting, and other types of content led by the speaker.
In our increasingly virtual work environment, Hour One's technology can also be optimally applied to e-learning, e-commerce and digital health, places where a human speaker is extremely valuable.From a company press release
The digital clone technology, according to Hour One, is already available. Sul company website there are numerous examples. While the initial iteration of the technology may seem slightly less than the real thing, it is already quite close. I don't dare to think about possible improvements and applications, for example in virtual reality environments.