Google has begun rolling out new artificial intelligence (AI) functionalities that enable interaction via voice commands with the Gmail, Google Docs, and Google Keep applications. These tools utilize Gemini models so users can ask questions, generate documents, and organize notes using natural language.
These features, named Gmail Live, Docs Live, and Keep Live, were first introduced in May during the Google I/O 2026 event. They are currently available to subscribers of Google AI plans, starting in English and on mobile devices running Android and iOS. Google also announced that these features will gradually be offered to Workspace corporate clients.
Specific application features
In Google Docs, the feature is designed to act as a writing collaborator. Instead of starting production on a blank screen, the user can verbally describe what they want to create, allowing the AI to structure ideas and prepare an initial draft. This functionality assists in drafting various types of documents, such as proposals and summaries, and also allows the user to request adjustments and add information during the interaction.
A notable aspect of Docs Live is its ability to access data from other Google services. With user permission, Gemini can consult relevant information from Gmail, Google Drive, Google Chat, and the web itself to enrich the document. The intention is to facilitate smoother communication, especially during brainstorming sessions, where the AI converts spoken content into cohesive text.
Docs Live is accessible to subscribers of Google AI Pro and Ultra plans. Google Keep has received Keep Live, which focuses on quickly recording ideas without the need for typing. The user can simply dictate what they are thinking, even if the thoughts are disorganized, and the AI transforms them into a structured note. Furthermore, the tool can generate lists from voice commands; for example, when requesting a list of ingredients for shakshuka, Keep Live automatically organizes this information.
The purpose of Keep Live is to convert the application into a type of voice scratchpad, where the individual can perform a 'brain dump' and delegate the subsequent organization of the material to the AI.
These innovations are part of Google's broader strategy to establish voice as a primary method of interaction with its products. Previously, in April, the company introduced cross-application dictation features in the Mac version. More recently, Google launched Rambler, a Gemini-based dictation tool capable of eliminating filler words from speech, and in August, announced Gemini 3.5 Transcribe, a model dedicated to speech-to-text conversion.
Unlike merely transcribing what is said, the new features in Gmail, Docs, and Keep propose a true conversation with the applications. The AI is expected to interpret requests, search for data, organize content, and execute tasks based on natural language. For now, Gmail Live, Docs Live, and Keep Live operate only in English, and Google confirmed that the rollout to eligible plans will occur that week, with expansion to Workspace business clients happening later.
