Automated call classification and screening

By generating and grouping call embeddings, and utilizing machine learning models to automatically answer and terminate spam calls, the problem of spam call identification is solved, enabling automatic detection on client devices and reducing the burden on users.

CN114667516BActive Publication Date: 2026-03-24GOOGLE LLC
View PDF -1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-11-18
Publication Date
2026-03-24

AI Technical Summary

Technical Problem

Spam calls, especially bot calls, are numerous and difficult to identify effectively using number-based anti-spam tools. Spoofed numbers further complicate the identification process.

Method used

Call embeddings are generated using a trained machine learning model, grouped into clusters, and cluster scores are determined based on representative embeddings and counts. Client devices automatically answer and terminate spam calls, and embeddings generated from call content are compared.

Benefits of technology

It effectively reduces user interaction with spam calls, alleviates the user burden, and enables automatic detection and termination of spam calls on client devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114667516B_ABST
    Figure CN114667516B_ABST
Patent Text Reader

Abstract

Implementations described herein relate to methods, systems, and computer-readable media for automatically answering calls. In some implementations, a method includes receiving, at a client device, a call from a calling device. The method also includes determining, based on an identifier related to the call, whether the call matches an automatic answer criteria, and in response to determining that the call matches the automatic answer criteria, answering the call without user input and without prompting a user of the client device. The method further includes generating a call embedding for the call based on audio received for the call, comparing the call embedding to a spam embedding to determine whether the call is a spam call, and in response to determining that the call is a spam call, terminating the call.
Need to check novelty before this filing date? Find Prior Art