Back to results
US 2023/0419367 A1Published application
Apparatus and method for communicating with users
AssigneeOne Call to All Limited
InventorChris Smith
Life of this patent
Estimating the term…
Abstract
An apparatus and method for communicating with users includes an automated communication platform connected to a network. Automated communication platform includes an originator and target users. Automated communication platform allows an originator to customize a message and send the message to a set of users, determined by the originator. The users may respond to a custom message using a communication ending. A filter may be used to filter out qualified users.
Claims20
An apparatus for communicating with users, the apparatus comprising:
- at least a processor communicatively connected to a network;
- and
- a memory communicatively connected to the processor, the memory containing instructions configuring the at least a processor to:
- connect to an automated communication platform, wherein the automated communication platform allows the processor to communicate with:
- an originator device;
- and
- at least a user device each associated with a user of a plurality of users;
- initiate a communication campaign on the automated communication platform, wherein initiating the communication campaign comprises:
- customizing a message comprising a communication ending;
- and
- choosing a user device of the at least a user device to receive the communication campaign;
- receive training data correlating filters and communication campaigns, wherein the training data iteratively trains a machine learning model that outputs custom filters, wherein training the machine learning model comprises:
- generating a plurality of first vector outputs, each first vector output containing a training data cluster representing a feature category of a filter;
- generating a second vector output containing a custom message;
- and
- calculating a distance between the first vector output and the second vector, wherein the first vector and second vector is normalized in order to make a vector comparison independent of absolute quantities of attributes;
- classify, using the machine learning model, a custom message to a custom filter for the communication campaign based on a calculated distance of equivalent similarity of the second vector output to a first vector output of the plurality of first vector outputs;
- import a file comprising a set of contact data associated with the chosen user device of the at least a user device;
- and
- receive an interaction from a user associated with the chosen user device of the at least a user device as a result of the communication campaign, wherein the interaction comprises a response to the custom filter embedded in the communication ending and receiving the interaction from the chosen user comprises:
- authenticating an identity of the user associated with the chosen user device as a function of a automatic speech recognition process and the interaction;
- and
- parsing the interaction for at least one keyword as a function of audio visual speech recognition comprising an audio-based speech recognition process and an image-based speech recognition process.
20 more elements
(canceled)
(canceled)
A method of communicating with users, the method comprising:
- connecting, by a processor, to an automated communication platform, wherein the automated communication platform allows the processor to communicate with:
- an originator device;
- and
- at least a user device each associated with a user of a plurality of users;
- initiating, by the processor, a communication campaign on the automated communication platform wherein initiating the communication campaign comprises:
- customizing a message comprising a communication ending;
- and
- choosing a user device of the at least a user device to receive the communication campaign;
- receiving training data correlating filters and communication campaigns, wherein the training data iteratively trains a machine learning model that outputs custom filters, wherein training the machine learning model comprises:
- generating a plurality of first vector outputs, each first vector output containing a training data cluster representing a feature category of a filter;
- generating a second vector output containing a custom message;
- and
- calculating a distance between the first vector output and the second vector, wherein the first vector and second vector is normalized in order to make a vector comparison independent of absolute quantities of attributes:
- classifying, using the machine learning model, a custom message to a custom filter for the communication campaign based on a calculated distance of equivalent similarity of the second vector output to a first vector output of the plurality of first vector outputs;
- importing, by the processor, a file comprising a set of contact data associated with the chosen user device of the at least a user device;
- and
- receiving, by the processor, an interaction from a user associated with the chosen user device of the at least a user device as a result of the communication campaign, wherein the interaction comprises a response to the custom filter embedded in the communication ending and receiving the interaction from the chosen user comprises:
- authenticating an identity of the user associated with the chosen user device as a function of at a automatic speech recognition process and the interaction;
- and
- parsing the interaction for at least one keyword as a function of audio visual speech recognition comprising an audio-based speech recognition process and an image-based speech recognition process.
17 more elements
(canceled)
(canceled)
Dependencies are inferred from the claim text.