Abstract:
Google is further upgrading Gemini Live from an assistant mainly used for natural voice conversations to a tool for productivity scenarios and intelligent agent workflows. Gemini Live, first launched in 2024, is designed to allow users to communicate with AI assistants in a more natural voice. Since then, Google has continued to expand its capabilities, including supporting more than 40 languages and linking with multiple Google apps. The latest upgrade focuses on strengthening task execution, cross-application collaboration and personalized service capabilities.
The most important change this time is that Gemini Live is connected to Google's intelligent agent system Spark. Users can now describe their needs directly through speech, and Spark continues to process complex multi-step tasks in the background across services such as Google Docs, Google Sheets, Google Drive, and the web.
For example, users can dictate scattered thoughts to Gemini Live and ask Spark to organize them into a Google Docs document outline; they can also have Gemini maintain a weekly meal menu and automatically generate a shopping list based on recipes saved in Docs.
Gemini Live has also added a new feature called "Daily Brief". Users can simply ask "What's my daily briefing?" and get a voice summary that consolidates important information from their Gmail messages and Google Calendar.
At the same time, Google has also added the ability to operate Gmail hands-free to Gemini Live. Users can use voice commands to search and summarize email content, as well as star, archive or delete emails.
In previous versions, each conversation in Gemini Live was almost independent and could not inherit the context of past exchanges. After the update, Gemini Live combines the content of previous conversations with information from connected services such as Gmail, Google Photos, Google Search, and YouTube to provide more personalized responses.
For example, Google said that Gemini Live can first summarize recent arrangements involved in Gmail, and then hand over subsequent tasks to Spark for processing, such as creating corresponding schedules in Google Calendar and adding estimated driving times.
In terms of subscription, using the Gemini Spark function requires opening a Google AI Pro or higher level subscription service; the "Daily Briefing" function requires users to subscribe to at least Google AI Plus.
Comments