Capture a spoken to-do
Turn a spoken reminder into a task the person can check before adding it to a list.
Illustrative idea, not a published runnable template. Grunt is prelaunch; live service is paused and live provider connections and outcomes remain unverified. The steps, providers, and outcomes below describe a concept.
Speak a request when typing is inconvenient and receive a response in text or aloud. The user can choose what context to share and what to do with the result; anything that changes a record, contacts another person, or commits the user to an action stays subject to their review and approval.
Actual availability depends on published steps, supported connections, permissions, and provider approval.
“Understand the user’s spoken request, prepare a useful response, and return it for the person to review before any consequential action is taken.”
Clear tasks, with review where it matters.
Helps people handle everyday requests by speaking naturally and reviewing a response on their device.
Speech recognition can mishear names, amounts, places, or instructions, and extra context such as location or images can be sensitive. Confirm the transcript and recipient before sharing, limit personal details to what the user intends to provide, and require explicit approval before sending messages or changing records.
Select an illustrative task to explore its proposed role. No workflow runs here.
Tell the assistant what help is needed and provide only the context necessary for the request.
Review the transcribed words and any optional information, correcting names, places, or details that were misheard.
Read or listen to the prepared answer and ask for clarification when it is incomplete or uncertain.
Decide whether to keep, copy, or otherwise use the result, and confirm the destination before sharing information.
Review the exact message, record change, or commitment and give clear approval before it is carried out.
Tell the assistant what help is needed and provide only the context necessary for the request.
These are planning notes for the concept, not instructions for a currently available template.
Decide which everyday questions the assistant may handle and which topics must be referred to a person.
Explain what voice, location, images, or files may be shared and who can access or retain that information.
Define which responses are informational and require a separate confirmation for messages, record changes, or other external effects.
Test an ordinary request, a misheard detail, and a consequential action to confirm the user can correct or decline the result.
Published templates and verified customer availability remain pending.
Explore the sample experience while launch details are being finalized.
View pricing status Explore the simulation