Google enhances Gemini with canvas, audio overview features: Check details

Google said that the Audio Overview feature converts documents, slides, and reports generated in Deep Research mode into an AI-generated audio with podcast-like discussion

Google has introduced new AI-powered features in its Gemini chatbot, aimed at enhancing productivity and workflow. These include Audio Overview, which generates podcast-style audio discussions from uploaded files, and Canvas, an interactive workspace for real-time collaboration on documents and code.

 

Audio Overview in Gemini

 

Google said that the Audio Overview feature converts documents, slides, and reports generated in Deep Research mode into an AI-generated audio with a podcast-like discussion. The tool creates a dialogue between two AI hosts, summarising the uploaded content while drawing connections between key topics. Google said that this feature aims to provide unique perspectives on a subject.

 

To use Audio Overview, users can upload a document or slides to Gemini and click the suggestion chip that appears above the prompt bar. The feature is available on both the web and mobile versions of Gemini, allowing users to share and download the generated audio.

Currently, Audio Overview is rolling out for Gemini and Gemini Advanced subscribers in English, with support for more languages coming soon.

 

Canvas in Gemini

 

As per Google, the Canvas feature is an interactive workspace within Gemini, designed for collaborative projects with the AI chatbot. It allows users to write, edit, and refine documents or code in real time. Unlike traditional chat responses, Gemini in Canvas mode can make targeted changes to specific sections, adjusting tone, length, and formatting as per user requests.

 

For developers, Canvas streamlines coding workflows by helping create initial prototypes for web apps, Python scripts, games, and other interactive applications. Gemini can also generate and preview HTML/React code, offering a visual representation of the user's design.

 

The Canvas interface is available to Gemini and Gemini Advanced subscribers in all supported languages. Users can activate it via the "Canvas" option in the prompt bar. However, Canvas currently works only with stable Gemini models like 2.0 Flash and does not support experimental models like 2.0 Pro.