Google is testing voice cloning features for its AI Studio, indicated by a hidden "Create Your Voice" option linked to the Flash native audio preview model associated with Gemini 2.5. Though currently inactive, this feature allows users to record and upload audio, hinting at a future release of Gemini 3 Flash with voice cloning capabilities. If launched, developers could create synthetic voices based on samples they provide, enhancing the personalization of AI-generated audio responses. This will be particularly useful for app developers focusing on assistants or accessibility tools. In addition to voice cloning, a new "Import from GitHub" option has surfaced in the build section. This will let users import entire repositories into AI Studio, streamlining the transition for developers moving projects into Google’s platform. However, specifics on supported frameworks and integration details are still unclear. Other planned enhancements include better backend capabilities like database and Firebase integration. The homepage of AI Studio is also undergoing changes, with updates to activity summaries and usage stats, as well as the introduction of comment panels. These adjustments signal Google’s commitment to evolving AI Studio into a more robust, developer-centric platform as the Gemini models progress. The rapid pace of these updates suggests that some features might be rolled out soon, depending on testing and stability.