/Patent Tracker
GoogleGranted patentAI & softwareInterest 7/10

Google patents live video chat that asks clarifying questions

Google has been granted a patent for a system that handles live video queries by using an orchestration model to decide when to ask users for more details. The system processes the original video input along with the user's real-time answer to generate a final response.

Why it mattersThis approach allows video-based assistants to resolve ambiguity in real-time, potentially making interactions with multimodal AI more accurate and conversational.

In the filing’s own words

Provided is a system for providing a live query response session for responding to multimodal input queries. A computing system initiates a live query response session. While conducting the live query response session, the computing system receives a multimodal input query. The computing system determines, by an orchestration model, one or more information retrieval steps to acquire additional information associated with the multimodal input query, wherein the information retrieval steps include a clarifying question requesting additional information from the user. The computing system provides the clarifying question to the user computing device. The computing system receives a user response to the clarifying question in real time. The computing system generates model input based on the multimodal input query and the user response. The computing system processes the model input with a query response model to generate a model output.

What exactly they claim (claim 1)

1. A computer-implemented method for providing a live query response session, wherein the method comprises: initiating, by a computing system with one or more processors, a live query response session with a computing device associated with a user; while conducting the live query response session: receiving, by the computing system, a multimodal input query from the computing device, wherein the multimodal input query includes video content; determining, by an orchestration model executing on the computing system, one or more information retrieval steps to acquire additional information associated with the multimodal input query, wherein the information retrieval steps include a clarifying question requesting additional information from the user; providing, by the computing system, the clarifying question to the user computing device; receiving, by the computing system, a user response to the clarifying question, wherein the user response is received in real-time; generating, by the computing system, model input based on the multimodal input query and the user response; processing, by the computing system, the model input with a query response model to generate a model output based on the model input; transmitting, by the computing system, the model output for display at a user computing device; receiving, by the computing device, audio content from the user computing device; and determining, by the computing device, that the audio content contains user feedback with respect to the multimodal input query.

Facts

Company
Google (Google LLC)
Official title
System and methods for providing a live query response session
Patent number
US12737419B2
Application number
19003352
Filed
Dec 27, 2024
Granted
Sep 15, 2026
Inventors
8
Claims
19

Classified under

G06F 16/735 Physics & computingG06F 16/732 Physics & computingG06F 16/738 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.