# US 2012/0253801 A1 · Automatic determination of and response to a topic of a conversation

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

- Number: US 2012/0253801 A1
- Document ID: US20120253801A1
- Status: Published application
- Kind: A1
- Application number: 13/073,579
- Filed: 2011-03-28
- Published: 2012-10-04
- Earliest claimed: 2011-03-28
- Assignees: Epic Systems Corporation
- Inventors: Chris Santos-Lang; Sumit Rana; Binit Mohanty; Rajeev Gangwar
- Web page: https://patentgrep.com/patent/US20120253801A1

## CPC

- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A system, computer-readable medium, and method for automatically determining a topic of a conversation and responding to the topic determination are provided. In the method, an active topic is defined as a first topic in response to execution of an application. The first topic includes first text defining a plurality of phrases, a probability of occurrence associated with each of the plurality of phrases, and a response associated with each of the plurality of phrases. Speech text recognized from a recorded audio signal is received. Recognition of the speech text is based at least partially on the probability of occurrence associated with each of the plurality of phrases of the first topic. A phrase of the plurality of phrases associated with the received speech text is identified. The response associated with the identified phrase is performed by the computing device. The response includes instructions defining an action triggered by occurrence of the received speech text, wherein the action includes defining the active topic as a second topic.

## Claims

### Claim 1

A system comprising:
a processor; and
a computer-readable medium operably coupled to the processor, the computer-readable medium having computer-readable instructions stored thereon that, when executed by the processor, cause the system to
in response to execution of an application by the system, define an active topic as a first topic, wherein the first topic includes first text defining a plurality of phrases, a probability of occurrence associated with each of the plurality of phrases, and a response associated with each of the plurality of phrases;
receive speech text recognized from a recorded audio signal, wherein recognition of the speech text is based at least partially on the probability of occurrence associated with each of the plurality of phrases of the first topic;
identify a phrase of the plurality of phrases associated with the received speech text; and
perform the response associated with the identified phrase, wherein the response includes instructions defining an action triggered by occurrence of the received speech text, and further wherein the action includes defining the active topic as a second topic, wherein the second topic includes second text defining a second plurality of phrases, a second probability of occurrence associated with each of the second plurality of phrases, and a second response associated with each of the second plurality of phrases.

### Claim 2 (depends on 1)

The system of claim 1, further comprising a microphone to generate the audio signal.

### Claim 3

A non-transitory computer-readable medium having stored thereon computer-readable instructions that when executed by a computing device cause the computing device to:
define an active topic as a first topic in response to execution of an application by the computing device, wherein the first topic includes first text defining a plurality of phrases, a probability of occurrence associated with each of the plurality of phrases based on the first topic, and a response associated with each of the plurality of phrases;
receive speech text recognized from a recorded audio signal, wherein recognition of the speech text is based at least partially on the probability of occurrence associated with each of the plurality of phrases of the first topic;
identify a phrase of the plurality of phrases associated with the received speech text; and
perform the response associated with the identified phrase, wherein the response includes instructions defining an action triggered by occurrence of the received speech text, and further wherein the action includes defining the active topic as a second topic, wherein the second topic includes second text defining a second plurality of phrases, a second probability of occurrence associated with each of the second plurality of phrases, and a second response associated with each of the second plurality of phrases.

### Claim 4 (depends on 3)

The computer-readable medium of claim 3, wherein the application includes input controls for accessing a plurality of user interface windows and for entering information in a database.

### Claim 5 (depends on 4)

The computer-readable medium of claim 4, wherein the application maintains electronic medical records in the database.

### Claim 6 (depends on 4)

The computer-readable medium of claim 4, wherein the action includes entering at least a portion of the speech text in a data entry field of a user interface window of the plurality of user interface windows.

### Claim 7 (depends on 4)

The computer-readable medium of claim 4, wherein the action includes presenting a user interface window of the plurality of user interface windows in a display operably coupled with the computing device.

### Claim 8 (depends on 7)

The computer-readable medium of claim 7, wherein the user interface window includes an image associated with the second topic.

### Claim 9 (depends on 7)

The computer-readable medium of claim 7, wherein the user interface window includes a video associated with the second topic.

### Claim 10 (depends on 4)

The computer-readable medium of claim 4, wherein the action includes changing a focus of a user interface window of the plurality of user interface windows presented in a display operably coupled with the computing device.

### Claim 11 (depends on 3)

The computer-readable medium of claim 3, wherein the action includes triggering execution of a second application by the computing device.

### Claim 12 (depends on 3)

The computer-readable medium of claim 3, wherein the first topic further includes a speaker associated with each of the plurality of phrases, wherein the probability of occurrence is further associated with the speaker.

### Claim 13 (depends on 3)

The computer-readable medium of claim 3, wherein the instructions include a plurality of actions triggered by occurrence of the received speech text.

### Claim 14 (depends on 3)

The computer-readable medium of claim 3, wherein the response includes an identifier of a response table, and further wherein the response table includes a plurality of sets of instructions.

### Claim 15 (depends on 14)

The computer-readable medium of claim 14, wherein the response table further includes a weight associated with each set of instructions of the plurality of sets of instructions.

### Claim 16 (depends on 15)

The computer-readable medium of claim 15, wherein the response table further includes a modifier associated with a set of instructions of the plurality of sets of instructions, and further wherein the modifier includes a secondary characteristic and a numerical adjustment value, wherein if the secondary characteristic is true, the weight associated with the set of instructions is adjusted using the numerical adjustment value

### Claim 17 (depends on 16)

The computer-readable medium of claim 16, wherein the set of instructions of the plurality of sets of instructions having the highest modified weight is performed.

### Claim 18 (depends on 3)

The computer-readable medium of claim 3, wherein the computer-readable instructions further cause the computing device to:
receive an indicator of an interaction with the application by a user;
in response to receipt of the indicator, determine if the second topic is a correct topic based on the interaction with the application; and
if the second topic is determined to be the correct topic based on the interaction with the application, increase the probability of occurrence associated with the identified phrase.

### Claim 19 (depends on 3)

The computer-readable medium of claim 3, wherein the computer-readable instructions further cause the computing device to:
receive an indicator of an interaction with the application by a user;
in response to receipt of the indicator, determine if the second topic is a correct topic based on the interaction with the application; and
if the second topic is determined to not be the correct topic based on the interaction with the application, decrease the probability of occurrence associated with the identified phrase.

### Claim 20 (depends on 19)

The computer-readable medium of claim 19, wherein the response includes an identifier of a response table, and further wherein the response table includes a plurality of sets of instructions.

### Claim 21 (depends on 20)

The computer-readable medium of claim 20, wherein, if the second topic is determined to not be the correct topic based on the interaction with the application, the computer-readable instructions further cause the computing device to select a different set of instructions of the plurality of sets of instructions, wherein the selected different set of instructions defines the active topic as a third topic.

### Claim 22 (depends on 20)

The computer-readable medium of claim 20, wherein, if the second topic is determined to not be the correct topic based on the interaction with the application, the computer-readable instructions further cause the computing device to:
determine if a different set of instructions of the plurality of sets of instructions can be selected; and
if a different set of instructions of the plurality of sets of instructions cannot be selected, remove the identified phrase from the first topic.

### Claim 23

A method of automatically determining a topic of a conversation and responding to the topic determination, the method comprising:
defining, by a computing device, an active topic as a first topic in response to execution of an application by the computing device, wherein the first topic includes first text defining a plurality of phrases, a probability of occurrence associated with each of the plurality of phrases, and a response associated with each of the plurality of phrases;
receiving speech text recognized from a recorded audio signal at the computing device, wherein recognition of the speech text is based at least partially on the probability of occurrence associated with each of the plurality of phrases of the first topic;
identifying a phrase of the plurality of phrases associated with the received speech text by the computing device; and
performing the response associated with the identified phrase by the computing device, wherein the response includes instructions defining an action triggered by occurrence of the received speech text, and further wherein the action includes defining the active topic as a second topic, wherein the second topic includes second text defining a second plurality of phrases, a second probability of occurrence associated with each of the second plurality of phrases, and a second response associated with each of the second plurality of phrases.
