Systems and methods for conversations with devices about media using interruptions and changes of subjects

• US 10,311,862 B2
• Filed: 12/23/2015
• Issued: 06/04/2019
• Est. Priority Date: 12/23/2015
• Status: Active Grant

• Current Assignee: Rovi Guides, Inc. (Adeia Inc.)
• Original Assignee: Rovi Guides, Inc. (Adeia Inc.)
• Inventors: Dawes, Charles, Klappert, Walter R.
• Primary Examiner(s): Colucci, Michael C
• Application Number:
  US14/757,910
• Publication Number:
  US 20170186425A1
• Time in Patent Office: 1,259 Days
• Field of Search: 704257, 704 9, 704 8, 704275, 704274, 704254, 707759
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/222 Barge in, i.e. overridable guidance for interrupting prompts

Abstract

Systems and methods are described herein for providing media guidance. Control circuitry may receive a first voice input and access a database of topics to identify a first topic associated with the first voice input. A user interface may generate a first response to the first voice input, and subsequent to generating the first response, the control circuitry may receive a second voice input. The control circuitry may determine a match between the second voice input and an interruption input such as a period of silence or a keyword or a phrase, such as “Ahh,”, “Umm,”, or “Hmm.” The user interface may generate a second response that is associated with a second topic related to the first topic. By interrupting the conversation and changing the subject from time to time, media guidance systems can appear to be more intelligent and human.

Claims

1. A method for mimicking human interactions in computer systems that use interactive vocal conversations when providing media guidance by automatically interrupting and changing subjects during the interactive vocal conversation in response to detecting a user hesitation, the method comprising:

receiving, using control circuitry, a first voice input;

accessing, using the control circuitry, a database of topics, the database of topics comprising a semantic network indicating relationships between a plurality of topics;

identifying, using the control circuitry, a first topic from the database of topics that is associated with the first voice input;

generating, using a user interface, a first response to the first voice input comprising a media asset recommendation associated with the first topic;

subsequent to generating the first response, receiving, using the control circuitry, a second voice input followed by a period of silence;

determining that the period of silence exceeds a threshold period of silence;

in response to determining that the period of silence exceeds the threshold period of silence, comparing the second voice input to a list of interruption inputs to determine a match between the second voice input and an interruption input from the list of interruption inputs, wherein the list of interruption inputs corresponds to a list of verbal cues indicative of hesitations; and

in response to determining the match indicative of a hesitation between the second voice input and the interruption input, generating, using the user interface, a second response to the first voice input comprising a media asset recommendation associated with a second topic from the database of topics that is associated with the first topic.

2. The method of claim 1, further comprising:

extracting, from the database, relationships between the first topic and a remainder of the plurality of topics;

comparing each of the relationships between the first topic and the remainder of the plurality of topics to a relationship threshold; and

storing to memory a list indicating a subset of the relationships between the first topic and the remainder of the plurality of topics that do not exceed the relationship threshold and a list of topics of the plurality of topics that correspond to the subset of the relationships;

wherein the second topic is selected from the list of topics.

3. The method of claim 1, wherein the semantic network comprises numerical relationships between the plurality of topics, the numerical relationships indicating a statistical likelihood that the second topic is related to the first topic.

4. The method of claim 3, further comprising:

receiving a plurality of voice inputs from a plurality of users;

identifying a first subset of the plurality of voice inputs that relate to the first topic;

identifying, from the first subset of the plurality of voice inputs, a second subset of the plurality of voice inputs relating to the second topic; and

calculating a statistical likelihood, based on the identified first subset of the plurality of voice inputs and second subset of the plurality of voice inputs, that the second topic follows the first topic.

5. The method of claim 1, further comprising:

accessing, using the control circuitry, a user profile indicating media preferences of a user;

retrieving a genre preference from the user profile;

identifying a subset of the plurality of topics that are associated with the retrieved genre; and

selecting the second topic from the subset of the plurality of topics.

6. The method of claim 1, wherein the database of topics indicates, for each of the plurality of topics, a genre associated with each respective topic, and further comprising:

extracting, from the database of topics, a genre associated with the first topic;

identifying a subset of the plurality of topics that are associated with the extracted genre; and

selecting the second topic from the subset of the plurality of topics.

7. The method of claim 1, further comprising:

receiving, using the control circuitry, a third voice input;

comparing the third voice input to the list of interruption inputs to determine a match between the third voice input and a second interruption input from the list of interruption inputs;

determining whether a threshold period of time has elapsed between a current time and the second response; and

in response to determining that the threshold period of time has elapsed between the current time and the second response, generating, using the user interface, a third response to the first voice input, wherein the third response is associated with a third topic from the database of topics that is associated with the first topic.

8. The method of claim 1, further comprising:

receiving an indication of a user associated with the first voice input;

accessing, using the control circuitry, a plurality of voice personality profiles, each voice personality profile corresponding to a respective user and comprising indications of a plurality of interruption inputs; and

selecting one of the plurality of voice personality profiles based on the indication of the user associated with the first voice input;

wherein comparing the second voice input to the list of interruption inputs comprises comparing the second voice input to the indications of the plurality of interruption inputs associated with the selected voice personality profiles.

9. The method of claim 8, further comprising:

extracting a threshold period of time from the selected voice personality profile; and

calculating a time elapsed since the first voice input by comparing a current time to a time associated with the first voice input;

wherein generating the second response is performed in response to determining that the time elapsed has exceeded the threshold period of time.

10. A system for mimicking human interactions in computer systems that use interactive vocal conversations when providing media guidance by automatically interrupting and changing subjects during the interactive vocal conversation in response to detecting user hesitations, the system comprising:

detection circuitry configured to receive a first voice input and a second voice input;

a user interface configured to generate a first response to the first voice input, comprising a media asset recommendation associated with a first topic, and a second response, comprising a media asset recommendation associated with the first topic, to the first voice input, wherein the second voice input is followed by a period of silence and is received subsequent to the first response;

a database of topics comprising a semantic network indicating relationships between a plurality of topics; and

control circuitry configured to;

access the database of topics;

identify the first topic from the database of topics that is associated with the first voice input; and

determine that the period of silence exceeds a threshold period of silence;

compare the second voice input to a list of interruption inputs in response to determining that the period of silence exceeds the threshold period of silence, wherein the list of interruption inputs corresponds to a list of verbal cues indicative of hesitations, to determine a match indicative of a hesitation between the second voice input and an interruption input from the list of interruption inputs; and

wherein the second response is generated by the user interface in response to determining the match between the second voice input and the interruption input, and wherein the second response is associated with the second topic from the database of topics that is associated with the first topic.

11. The system of claim 10, wherein the control circuitry is further configured to:

extract, from the database, relationships between the first topic and a remainder of the plurality of topics;

compare each of the relationships between the first topic and the remainder of the plurality of topics to a relationship threshold; and

store to memory a list indicating a subset of the relationships between the first topic and the remainder of the plurality of topics that do not exceed the relationship threshold and a list of topics of the plurality of topics that correspond to the subset of the relationships;

wherein the second topic is selected from the list of topics.

12. The system of claim 10, wherein the semantic network comprises numerical relationships between the plurality of topics, the numerical relationships indicating a statistical likelihood that the second topic is related to the first topic.

13. The system of claim 12, wherein the detection circuitry is further configured to:

receive a plurality of voice inputs from a plurality of users; and

wherein the control circuitry is further configured to;

identify a first subset of the plurality of voice inputs that relate to the first topic;

identify, from the first subset of the plurality of voice inputs, a second subset of the plurality of voice inputs relating to the second topic; and

calculate a statistical likelihood, based on the identified first subset of the plurality of voice inputs and second subset of the plurality of voice inputs, that the second topic follows the first topic.

14. The system of claim 10, wherein the control circuitry is further configured to:

access a user profile indicating media preferences of a user;

retrieve a genre preference from the user profile;

identify a subset of the plurality of topics that are associated with the retrieved genre; and

select the second topic from the subset of the plurality of topics.

15. The system of claim 10, wherein the database of topics indicates, for each of the plurality of topics, a genre associated with each respective topic, and wherein the control circuitry is further configured to:

extract, from the database of topics, a genre associated with the first topic;

identify a subset of the plurality of topics that are associated with the extracted genre; and

select the second topic from the subset of the plurality of topics.

16. The system of claim 10, wherein the control circuitry is further configured to:

receive a third voice input;

compare the third voice input to the list of interruption inputs to determine a match between the third voice input and a second interruption input from the list of interruption inputs; and

determine whether a threshold period of time has elapsed between a current time and the second response; and

wherein the user interface is configured to, in response to the determination that the threshold period of time has elapsed between the current time and the second response, generating a third response to the first voice input, wherein the third response is associated with a third topic from the database of topics that is associated with the first topic.

17. The system of claim 10, wherein the control circuitry is further configured to:

receive an indication of a user associated with the first voice input;

access a plurality of voice personality profiles, each voice personality profile corresponding to a respective user and comprising indications of a plurality of interruption inputs; and

select one of the plurality of voice personality profiles based on the indication of the user associated with the first voice input;

wherein the control circuitry is configured to compare the second voice input to the list of interruption inputs by comparing the second voice input to the indications of the plurality of interruption inputs associated with the selected voice personality profiles.

18. The system of claim 17, wherein the control circuitry is further configured to:

extract a threshold period of time from the selected voice personality profile; and

calculate a time elapsed since the first voice input by comparing a current time to a time associated with the first voice input;

wherein the control circuitry is configured to generate the second response in response to determining that the time elapsed has exceeded the threshold period of time.