Yesterday, Google announced that it has updated its Gemini AI tool with two new additions; a Canvas feature, which allows users to fine-tune documents and code, and an Audio Overview feature that transforms documents into audio files similar to podcasts. These additions were shared in a blog post on Tuesday by Dave Citron, Senior Director of Product Management at Google.
On the programming side of things, Citron explained that the Canvas feature is meant to simplify how programmers code with Gemini as it obviates the need to switch between various applications. This feature helps ease the process of “transforming coding ideas into working prototypes for web apps, Python scripts, games, simulations and other interactive apps.”
According to Google, with Canvas, you can create “HTML/React code and other web prototypes” and immediately see how they will appear on your webpage. Providing insights on how it works using an email subscription form as an example, Citron explains that you can “ask Gemini to generate the HTML for the form and then preview how it will appear and function within your web app.” Afterward, you can adjust the form and share what you’ve done with others.
To use this feature, simply select Canvas on the prompt bar and input your instructions.
These discussions can apparently take the form of a summary of the material, a unique perspective on the issues, or an explanation of similarities between topics in the provided material or documents. To use this feature, you will need to upload the documents on the Audio Overview and click the “Generate Audio Overview button.”
Google has revealed that both the Canvas and the Audio Overview feature are available for Gemini and Gemini advanced users worldwide; however, the Audio Overview feature is only available in English with more languages to follow.