# US 2026/0292056 A1 · System and method for handling unwanted calls and messages

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0292056 A1
- Document ID: US20260292056A1
- Status: Published application
- Kind: A1
- Application number: 19/082,611
- Filed: 2025-03-18
- Published: 2026-09-24
- Earliest claimed: 2025-03-18
- Assignees: AT&T Intellectual Property I, L.P.; AT&T Mobility II LLC
- Applicants: AT&T Intellectual Property I, L.P.; AT&T Mobility II LLC
- Inventors: Karen Stansbury; Howard L. Lang; Derek Baylor; Antoine Diffloth; Arthur Dileo, JR.; Mitchell Gunnels; Phillip Weeks, JR.
- Web page: https://patentgrep.com/patent/US20260292056A1

## CPC

- H04M 3/2281: {Call monitoring, e.g. for law enforcement purposes; Call tracing; Detection or prevention of malicious calls}
- H04M 1/72433: for voice messaging, e.g. dictaphones (for answering incoming calls H04M1/64)
- H04M 1/72436: for text messaging, e.g. short messaging services [SMS] or e-mails
- H04M 1/72484: wherein functions are triggered by incoming communication events
- H04M 3/5335: {Message type or catagory, e.g. priority, indication}
- H04M 2201/39: using speech synthesis
- H04M 2201/405: involving speaker-dependent recognition
- H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08)

## Abstract

Aspects of the subject disclosure may include, for example, receiving, via an application programming interface, a communication content forwarded from a mobile device, wherein the communication content is selected from the group consisting of a voicemail message, an active call audio segment, and a text message, analyzing the communication content, wherein the analyzing comprises analyzing the communication content with artificial intelligence (AI) models and audio signal processing to determine an intent of the communication content and a probability that the communication content is AI generated, synthetic, or cloned, generating a threat indication based on the intent of the communication content and a probability that the communication content is AI generated, synthetic, or cloned, and transmitting the threat indication to the mobile device for display to a subscriber. Other embodiments are disclosed.

## Claims

### Claim 1

A device, comprising:
a processing system including a processor; and
a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:
receiving a forwarded communication from a user device, wherein the forwarded communication is selected from an audio communication and a text communication;
responsive to receiving the text communication as the forwarded communication, analyzing the text communication to determine an intent of the forwarded communication;
responsive to receiving the text communication as the forwarded communication, determining whether the text communication contains artificial intelligence generation characteristics;
responsive to receiving the audio communication as the forwarded communication, analyzing audio features of the audio communication to determine if a voice of the forwarded communication is synthesized;
responsive to receiving the audio communication as the forwarded communication, determining whether the audio communication includes synthetic or cloned voice characteristics;
generating a threat level indicator based on the intent of the forwarded communication, a presence of artificial intelligence generation characteristics, and a presence of synthetic or cloned voice characteristics of the audio communication; and
providing the threat level indicator to the user device.

### Claim 2 (depends on 1)

The device of claim 1, wherein the receiving the forwarded communication comprises:
receiving a voicemail message or an active call audio segment received at the user device as the audio communication.

### Claim 3 (depends on 1)

The device of claim 1, wherein the operations further comprise:
receiving a bridge of an active call between the user device and a network server;
receiving an active call audio segment of the active call at the network server;
analyzing an audio signature of the active call audio segment to determine the intent of the active call audio segment;
analyzing the audio signature of the active call audio segment to identify synthetic or cloned voice characteristics of the active call audio segment; and
providing, to the user device, a live indication of the threat level indicator, substantially in real time during the active call.

### Claim 4 (depends on 3)

The device of claim 3, wherein the receiving the bridge of an active call between the user device and a network server is responsive to a user actuation of a user interface of the user device and wherein the operations further comprise:
receiving full call audio of the active call at the network server following the bridge of the active call between the user device and the network server.

### Claim 5 (depends on 1)

The device of claim 1, wherein the operations further comprise:
receiving the forwarded communication from the user device at an application programming interface (API) of a network server, including
receiving text corresponding to a short message service (SMS) text message communicated to the user device on a mobile communication network; or
receiving text corresponding to an over-the-top text message communicated to an application operating on the user device over a communication network.

### Claim 6 (depends on 1)

The device of claim 1, wherein the operations further comprise:
receiving, from the user device, data corresponding to a screenshot image of a text conversation displayed on a display screen of the user device; and
using visual processing, extracting text of the screenshot image, forming the text communication.

### Claim 7 (depends on 1)

The device of claim 1, wherein the operations further comprise:
receiving data of a voicemail message from the user device at a network server;
transcribing the data of a voicemail message into transcribed text; and
analyzing the transcribed text to determine the intent of the voicemail message and a likelihood of a synthetic or cloned voice in the voicemail message.

### Claim 8 (depends on 1)

The device of claim 1, wherein the operations further comprise:
transcribing the audio communication into transcribed text; and
analyzing the transcribed text to determine the intent of the forwarded communication.

### Claim 9 (depends on 8)

The device of claim 8, wherein the operations further comprise:
analyzing the audio features of the audio communication using an ensemble of models to identify synthetic or cloned voice characteristics in the audio communication.

### Claim 10 (depends on 9)

The device of claim 9, wherein the analyzing the audio features of the audio communication using an ensemble of models comprises:
analyzing the audio features of the audio communication with at least one convolutional neural network module and one generative artificial intelligence module.

### Claim 11

A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:
receiving, via an application programming interface, a communication content forwarded from a mobile device, wherein the communication content is selected from the group consisting of a voicemail message, an active call audio segment, and a text message;
analyzing the communication content, wherein the analyzing comprises analyzing the communication content with artificial intelligence (AI) models and audio signal processing to determine an intent of the communication content and a probability that the communication content is AI generated, synthetic, or cloned;
generating a threat indication based on the intent of the communication content and a probability that the communication content is AI generated, synthetic, or cloned; and
transmitting the threat indication to the mobile device for display to a subscriber.

### Claim 12 (depends on 11)

The non-transitory machine-readable medium of claim 11, wherein the analyzing the communication content comprises:
analyzing audio features of the communication content using an ensemble of models to identify synthetic or cloned voice characteristics in the communication content.

### Claim 13 (depends on 12)

The non-transitory machine-readable medium of claim 12, wherein the analyzing the communication content comprises:
analyzing the communication content using a convolutional neural network for detecting synthetic voice characteristics of the communication content.

### Claim 14 (depends on 11)

The non-transitory machine-readable medium of claim 11, wherein the operations further comprise:
transcribing the voicemail message or the active call audio segment into transcribed text; and
analyzing the transcribed text to determine the intent of the communication content.

### Claim 15 (depends on 11)

The non-transitory machine-readable medium of claim 11, wherein the operations further comprise:
generating the threat indication when the probability that the communication content is AI generated, synthetic, or cloned exceeds a threshold value of 75%; and
transmitting the threat indication via push notification to the mobile device.

### Claim 16

A method,
comprising:
receiving, by a processing system including a processor of a network server in a communications network, a forwarded communication from a user device active on the communication network, wherein the forwarded communication is selected from an audio communication and a text communication;
analyzing, by the processing system, the forwarded communication to determine an intent of the forwarded communication, wherein the analyzing, when applied to the text communication, comprises determining whether the text communication contains artificial intelligence generation characteristics, and wherein the analyzing, when applied to the audio communication, comprises both analyzing audio features to determine an intent of the audio communication and determining whether the audio communication includes synthetic voice characteristics or cloned voice characteristics;
generating, by the processing system, a threat level indicator based on the intent of the forwarded communication and a presence of the artificial intelligence generation characteristics or the synthetic voice characteristics or cloned voice characteristics; and
providing, by the processing system, the threat level indicator to the user device.

### Claim 17 (depends on 16)

The method of claim 16, wherein the analyzing the forwarded communication comprises:
analyzing, by the processing system, the forwarded communication utilizing artificial intelligence (AI) models, including a deep learning algorithm, to enhance an accuracy of intent determination and detection of AI characteristics.

### Claim 18 (depends on 16)

The method of claim 16, wherein the analyzing the forwarded communication comprises:
analyzing, by the processing system, text communications utilizing natural language processing (NLP) to improve determination of intent.

### Claim 19 (depends on 16)

The method of claim 16, wherein the analyzing the forwarded communication comprises:
analyzing, by the processing system, contextual information, including ender identity information and historical communication patterns, to enhance accuracy of determination of intent.

### Claim 20 (depends on 16)

The method of claim 16, wherein the providing the threat level indicator to the user device comprises:
providing, by the processing system, the threat level indicator having different levels related to a perceived level of severity of a threat associated with the forwarded communication from the user device.
