Google has introduced another text-to-image tool called ImageFX. This tool lets users create images with text prompts. ImageFX distinguishes itself with its user-friendly interface. It features ‘expressive chips’ that helps users experiment with various dimensions of their creations and ideas.
Imagen 2 model powers ImageFX. This model helps ImageFX deliver the highest-quality AI-generated images.
Google has also made significant investments in Imagen 2 training data safety. The company has implemented guardrails to limit outputs featuring violent, offensive, or sexually explicit content. Additionally, filters have been applied to reduce the risk of generating images of named individuals.
Google has also announced improvements in MusicFX and TextFX. Upgrade in MusicLM brings faster generation of music and higher-quality audio. Now it can generate songs of up to seconds. TextFX update brings improved navigation and overall user experience.
Google has also created SynthID, a digital watermark, to confirm that AI created the content. It will also tag content made using ImageFX and MusicFX using SynthID. ImageFX content will also include IPTC metadata.
Starting today, users in the United States, Kenya, New Zealand, and Australia can explore the new and improved tools in the AI Test Kitchen, available exclusively in English.
