# US 11,790,904 B2 · Voice application platform

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

- Number: US 11,790,904 B2
- Document ID: US11790904B2
- Status: Granted
- Kind: B2
- Application number: 17/104,242
- Filed: 2020-11-25
- Published: 2021-10-21
- Granted: 2023-10-17
- Earliest claimed: 2018-06-05
- Assignees: Voicify, LLC
- Applicants: Voicify, LLC
- Inventors: Robert T. Naughton; Nicholas G. Laidlaw; Alexander M. Dunn; Jeffrey K. McMahon
- Web page: https://patentgrep.com/patent/US11790904B2

## CPC

- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)}
- G06F 16/9038: Presentation of query results
- G10L 15/02: Feature extraction for speech recognition; Selection of recognition unit
- G10L 15/28: Constructional details of speech recognition systems
- G10L 2015/221: {Announcement of recognition results}
- G10L 2015/223: {Execution procedure of a spoken command}
- G10L 2015/225: {Feedback of the input speech}
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Among other things, requests are received from voice assistant devices expressed in accordance with different corresponding protocols of one or more voice assistant frameworks. Each of the requests represents a voiced input by a user to the corresponding voice assistant device. The received requests are re-expressed in accordance with a common request protocol. Based on the received requests, responses to the requests are expressed in accordance with a common response protocol. Each of the responses is re-expressed according to a protocol of the framework with respect to which the corresponding request was expressed. The responses are sent to the voice assistant devices for presentation to the users.

## Claims

### Claim 1

A computer-implemented method comprising
defining a graph of nodes and edges to be traversed based on data derived from requests for service that conform to one or more different voice assistant frameworks and are based on speech of end-users expressing intents,
the nodes including an initial node at which to begin traversal of the graph to reach features that match respective requests for service conforming to the different voice assistant frameworks, and
adding a node to the graph to serve as the initial node of the graph at which to begin the traversal with respect to requests that conform to an additional voice assistant framework.

### Claim 2 (depends on 1)

The method of claim 1 in which voice assistant devices from which the requests are received conform to two or more different voice assistant frameworks.

### Claim 3 (depends on 1)

The method of claim 1 in which the data is derived from the requests for service by abstracting information in the requests to a data format that is common across the two or more different voice assistant frameworks.

### Claim 4 (depends on 1)

The method of claim 1 comprising updating nodes of the graph using outputs of a machine learning algorithm.

### Claim 5 (depends on 1)

The method of claim 1, comprising traversing the graph to formulate a response to a request message, wherein traversing the graph comprises:
determining the initial node based on an intent associated with the request message.

### Claim 6 (depends on 5)

The method of claim 5, wherein traversing the graph comprises:
determining one or more slots associated with the request message; and
determining an endpoint node to traverse to from the initial node, based on a number of common slot types shared between the one or more slots associated with the request message and one or more slots that share an edge with the endpoint node.

### Claim 7 (depends on 5)

The method of claim 5, wherein traversing the graph comprises:
determining, based on the request message, an endpoint node corresponding to a feature to be utilized to formulate the response to the request message;
determining a shortest path between the initial node and the endpoint node; and
traversing the graph using the determined shortest path.

### Claim 8 (depends on 5)

The method of claim 5, wherein traversing the graph comprises:
determining an endpoint node to traverse to from the initial node, wherein determining the endpoint node is based on a number of edges between the endpoint node and the initial node.

### Claim 9 (depends on 1)

The method of claim 1, wherein the graph comprises:
a first node corresponding to a feature utilized to formulate responses to request messages;
a second node sharing a first common edge with the first node, the second node associated with a first voice assistant framework; and
a third node sharing a second common edge with the first node, the third node associated with a second voice assistant framework.

### Claim 10

An apparatus comprising:
one or more processors; and
a storage storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
defining a graph of nodes and edges to be traversed based on data derived from requests for service that conform to one or more different voice assistant frameworks and are based on speech of end-users expressing intents,
the nodes including an initial node at which to begin traversal of the graph to reach features that match respective requests for service conforming to the different voice assistant frameworks, and
adding a node to the graph to serve as the initial node of the graph at which to begin the traversal with respect to requests that conform to an additional voice assistant framework.

### Claim 11 (depends on 10)

The apparatus of claim 10 in which voice assistant devices from which the requests are received conform to two or more different voice assistant frameworks.

### Claim 12 (depends on 10)

The apparatus of claim 10 in which the data is derived from the requests for service by abstracting information in the requests to a data format that is common across the two or more different voice assistant frameworks.

### Claim 13 (depends on 10)

The apparatus of claim 10, wherein the operations comprise updating nodes of the graph using outputs of a machine learning algorithm.

### Claim 14 (depends on 10)

The apparatus of claim 10, wherein the operations comprise traversing the graph to formulate a response to a request message, wherein traversing the graph comprises:
determining the initial node based on an intent associated with the request message.

### Claim 15 (depends on 14)

The apparatus of claim 14, wherein traversing the graph comprises:
determining one or more slots associated with the request message; and
determining an endpoint node to traverse to from the initial node, based on a number of common slot types shared between the one or more slots associated with the request message and one or more slots that share an edge with the endpoint node.

### Claim 16 (depends on 14)

The apparatus of claim 14, wherein traversing the graph comprises:
determining, based on the request message, an endpoint node corresponding to a feature to be utilized to formulate the response to the request message;
determining a shortest path between the initial node and the endpoint node; and
traversing the graph using the determined shortest path.

### Claim 17 (depends on 14)

The apparatus of claim 14, wherein traversing the graph comprises:
determining an endpoint node to traverse to from the initial node, wherein determining the endpoint node is based on a number of edges between the endpoint node and the initial node.

### Claim 18 (depends on 10)

The apparatus of claim 10, wherein the graph comprises:
a first node corresponding to a feature utilized to formulate response to request messages;
a second node sharing a first common edge with the first node, the second node associated with a first voice assistant framework; and
a third node sharing a second common edge with the first node, the third node associated with a second voice assistant framework.

### Claim 19

A non-transitory storage storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
defining a graph of nodes and edges to be traversed based on data derived from requests for service that conform to one or more different voice assistant frameworks and are based on speech of end-users expressing intents,
the nodes including an initial node at which to begin traversal of the graph to reach features that match respective requests for service conforming to the different voice assistant frameworks, and
adding a node to the graph to serve as the initial node of the graph at which to begin the traversal with respect to requests that conform to an additional voice assistant framework.

### Claim 20 (depends on 19)

The non-transitory storage of claim 19, in which voice assistant devices from which the requests are received conform to two or more different voice assistant frameworks.
