Google introduces a number of updates on the generative AI front, including a new text-to-image tool. What is the difference ImageFX that it has an interface that contains “expressive chips”. The idea is that these will help you “quickly experiment with adjacent dimensions of your creation and ideas.”
Along with the debut of ImageFX, Google says it’s improving MusicFX and TextFX. The company claims that it has made improvements to the MusicLM model that include faster music generation and higher quality audio, along with new features. Created songs can now last up to 70 seconds. As for TextFXGoogle has released usability updates to improve navigation and overall user experience.
Images created by ImageFX and audio created with MusicFX are tagged by SynthID, digital watermark It aims to identify fakes using AI, especially when they appear in Search or Chrome. ImageFX creations will also include IPTC metadata. According to Google, this will “give people more information when they encounter our AI-generated images.”
People in the US, Kenya, New Zealand, and Australia can try out these new and updated tools in the AI Test Kitchen starting today. They are currently only available in English.
Imagen 2 leverages ImageFX’s new image creation features. It’s also the technology driving the new generative AI options in Bard. Search, Ads, Duet AI and Vertex AI in Workspace. Google says that Imagen 2 helps deliver the highest quality AI-generated images yet. The company notes that the model helps clean up images of artifacts and improves areas of image creation where such tools have so far struggled.
In addition, Google says that Imagen 2 has made “significant investments” in the security of its training data, while “adding safeguards to limit problematic content such as violence, abuse, or violence.” explicit sexual content while applying filters to reduce the risk to create pictures of named persons.“This is due to improvements in the model’s ability to create photorealistic images. The company claims that it also performs “extensive competitor testing” to detect and limit potentially problematic and harmful content.
Elsewhere, Gemini Pro in Bard is more widely available starting today. this now available In over 40 languages and north of 230 countries and territories. Also, starting today, Google says people in most countries can create images Bard free in english. These images include SynthID watermarks.