Google's Gemini AI can now analyse your audio files: All you need to know

Google's Gemini app now lets Android, iOS, and web users upload audio files for AI to do a comprehensive analysis, opening new use cases like transcribing interviews, lectures, and voice memos

Gemini can now analyse your audio files. Google has enabled the option across platforms, including Android, iOS and web, for users to upload audio files for analysis on the Gemini app. The vice president of Google Labs and Gemini, Josh Woodward, announced this development on X (formerly Twitter). As per a Google support page, free-tier users can upload up to 10 minutes of audio and have five daily prompts.

 

This opens Google Gemini AI to new use cases – think transcribing interviews, analysing voice memos, or converting lecture audio into searchable AI dialogue. It lifts the app beyond typed or spoken prompts, offering multimodal input.

 

 

Along with this, Google has added support for five languages, including Hindi, to Search’s AI mode. Furthermore, as per Woodward’s reply to a comment on this post revealed that now Google is working to build an Auto mode in Gemini so that users won’t have to manually switch Gemini Flash and Pro mode. 

Limit for uploading audio files in Gemini

Free-tier users can upload up to 10 minutes of audio and have five daily prompts. Those subscribed to Gemini AI Pro or AI Ultra can use uploads up to three hours, across up to 10 files, even packaged within a single ZIP archive.

Expanding multilingual reach in Google Search

 

Also on Monday, Google enhanced its Search AI Mode by adding support for five new languages: Hindi, Indonesian, Japanese, Korean, and Brazilian Portuguese. Powered by Gemini 2.5 technology, this expansion allows more users globally to ask complex questions in their native tongues while using Search’s AI-driven tools to explore content. 

NotebookLM gets smarter

Furthermore, Google’s NotebookLM – a Gemini-powered research assistant – has gained the ability to generate structured reports such as blog posts, study guides, quizzes, and flashcards, all from uploaded documents or media. These are now available in over 80 languages. Users can select the desired style and tweak tone or format, with full rollout expected by end of week, according to a comment shared on X.

 

While audio upload is entirely new to the Gemini app, NotebookLM had already offered audio input, maintaining its positioning as a deep-dive research assistant rather than a conversational interface.