Google’s Gemini AI assistant now supports audio file uploads, allowing users to transcribe, summarize, and extract key information from recordings up to 10 minutes long. This feature is accessible through web and mobile apps via a user-friendly file-upload interface, distinguishing it from real-time voice command processing. Josh Woodward, VP of Gemini, noted that this function was heavily requested by users, indicating a demand for efficient audio processing.
During testing, Gemini effectively transcribed various audio types with minor errors and generated actionable insights like to-do lists. This enhancement aligns with Gemini’s ongoing integrations and personalization options. While Gemini’s audio capabilities are akin to those of competitors like ChatGPT and Anthropic’s Claude, it emphasizes everyday uses for a wider audience.
Currently, audio upload limitations include a 10-minute cap for recordings and daily quotas for free-tier users, which could hinder extensive processing needs. Users should manage their audio processing to avoid exceeding limitations.
Source link
