Yandex's "Alice AI" Learned to Describe What's Happening on Smart IP Cameras

Yandex added the ability to ask its proprietary voice assistant "Alice AI" about what's happening in the frame of smart IP cameras. The corresponding feature is already available to device owners.

How to ask Alice about what's on screen

The assistant accepts questions through any of the channels available to the user.

• Chat with "Alice AI"
• "Yandex Station" smart speaker
• "TV Station"
• "Yandex Drops" earbuds

Why Yandex needs a multimodal assistant

Previously, "Alice" primarily responded to voice and text queries, without "seeing" what was happening around the user. The ability to analyze video streams from home IP cameras adds visual perception to the assistant—a step toward multimodal AI helpers, which major technology companies worldwide are increasingly developing.

What this means

The feature transforms Yandex's smart cameras from simple video surveillance into a resource that can be queried by voice or text. This expands the "Alice AI" product line from speech and text recognition to analysis of visual information from home devices, moving the company's ecosystem closer to a multimodal assistant that can see what's happening around the user.

Frequently asked questions

What can "Alice AI" tell you about the camera feed?

The assistant answers questions about what is located or happening in the frame of a Yandex smart IP camera.

What devices can you use to ask Alice about the camera?

You can ask the question through a chat with "Alice AI", the "Yandex Station" smart speaker, "TV Station", or "Yandex Drops" earbuds.