Google Gemini adds audio uploads for transcripts and summaries

Google's Gemini app now accepts uploaded audio files on Android, iOS and the web, enabling users to transcribe, summarize and extract information from recordings, the company announced this week. The feature, long requested by users, lets people drop in common audio formats such as MP3, M4A and WAV through the app's plus menu or the web Upload files option and have Gemini analyze the content for searchable transcripts, speaker identification, action items and highlights.

According to Google's support documentation and an announcement from Josh Woodward, vice president of Google Labs and Gemini, free-tier users can upload up to 10 files per prompt with a combined total audio length capped at 10 minutes. Subscribers to Google AI Pro or Google AI Ultra have a higher limit of three hours of audio. The new audio capability joins existing multimodal support for images, PDFs and video; video uploads remain limited to five minutes for free users and one hour for paid subscribers, with a 2 gigabyte size ceiling for video files and a 100 megabyte limit for most other individual files.

Gemini's file-upload workflow also supports ZIP archives, code folders and GitHub repositories. Users can...