Google in turn unveils a tool that allows you to create videos from text
Google's Imagen Video tool can create five-second clips from a phrasing. This image was generated with the following words: a teddy bear running in New York. (Image generated by artificial intelligence).
The race for AI-generated content continues. A few days after the unveiling of Make-A-Video, a tool created by Meta that allows you to transform short sentences into video without sounds, Google responded on Wednesday by unveiling the first sequences of its Imagen Video software.
Like Make-A-Video, Imagen Video has the ability to create five-second videos from scratch from just a few words, like a panda taking a selfie.
Start of widget . Skip widget?End of widget. Back to top of widget?
To achieve this, Google trained its software through machine learning. It has been exposed to more than 60 million images linked to short descriptive texts, as well as some 14 million videos.
If the Google tool offers imperfect images, it is however able to generate images of text, a feature that its predecessors lacked.
The American giant also has another card in hand: Phenaki, a tool similar to Imagen Video, but which relies on the duration of the sequences at the expense of the quality of the images. This can generate videos up to 2.5 minutes long.
Here is an example of what Phenaki managed to create with the following sentences: Side view d& #x27;an astronaut walking in a puddle on Mars. Astronaut is dancing on Mars. Astronaut walks his dog on Mars. The astronaut and his dog are watching fireworks.
Start of the widget. Skip widget? End of widget . Back to top widget?
Google hopes to eventually join forces with its two tools, which are still under development. The company did not say whether a trial version of Imagen Video or Phenaki will be available to the public.
With information from TechCrunch, and PC Mag