Google reveals another text-to-image generative AI tool, ImageFX

is rolling out a series of updates on the generative AI front, including a new text-to-image conversion tool. What is different is that it has an interface with “expressive chips”. The idea here is that these will help you “quickly experiment with adjacent dimensions of your creation and ideas.”

Along with the debut of ImageFX, Google claims to have improved MusicFX and TextFX. The company says it has made upgrades to the MusicLM model that include faster music generation and higher quality sound, as well as new features. Generated songs can now be up to 70 seconds long. As for Google has rolled out usability updates with the aim of improving navigation and overall user experience.

Images generated by ImageFX and audio made with MusicFX are tagged by SynthID, which aims to clarify that these are forged using AI, especially when they appear in search or in Chrome. ImageFX creatives will also include IPTC metadata. According to Google, this will provide “people with more information every time they encounter our AI-generated images.”

People in the United States, Kenya, New Zealand, and Australia can try these new and revamped tools in AI Test Kitchen starting today. They are only available in English at the moment.

The Imagen 2 template powers ImageFX's new image generation features. It is also the technology that gives rise to new generative AI options in Bard, , Ads, Duet AI in Workspace and Vertex AI. Google says Imagen 2 helps deliver its highest quality AI-generated images to date. The company notes that the model helps keep images free of artifacts and improves areas of image generation that these tools have struggled with until now.

Additionally, Google says it has made “significant investments” in the security of Imagen 2 training data while adding safeguards to “limit problematic outputs such as violent, offensive, or offensive outputs.” as well as the application of filters to reduce the risk of This is due to the model's improved ability to generate photorealistic images. The company says it also conducts “extensive adversarial testing” to detect and crack down on potentially problematic and harmful content.

Somewhere else, is more widely available starting today. It is in more than 40 languages ​​and north of 230 countries and territories. Also today, Google says that users in most countries can generate images in in English for free. These images will include SynthID watermarks.

Source link

Leave a Comment