Personalized digital assistant device and related methods

• US 10,768,954 B2
• Filed: 01/30/2019
• Issued: 09/08/2020
• Est. Priority Date: 01/30/2018
• Status: Active Grant

• Current Assignee: Peloton Interactive, Inc.
• Original Assignee: Aiqudo, Inc.
• Inventors: Hemaraj, Kiran Bindhu, Mukherjee, Rajat
• Primary Examiner(s): Titcomb, William D
• Application Number:
  US16/261,657
• Publication Number:
  US 20190235887A1
• Time in Patent Office: 587 Days
• Field of Search: 715708
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G06F 9/453 Help systems
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Embodiments described herein are generally directed towards systems and methods relating to a crowd-sourced digital assistant system and techniques for disambiguating commands based on personalized usage of a digital assistant device, among other things. In various embodiments, the digital assistant device can use personal data, collected device usage data, and other types of collected contextual information, to disambiguate received commands for the proper selection and execution of operations on the digital assistant device. The digital assistant can process and interpret ambiguous commands and even unique user dialects without requiring extensive training to recognize and act on the received commands, even if the particular phraseology of the command has not previously been encountered by the digital assistant.

Claims

1. A computer-implemented method for disambiguating commands to perform digital assistant operations, comprising:

obtaining, by a digital assistant device, a command representation that corresponds to speech data received as input;

generating, by the digital assistant device, a query that includes the command representation to search a stored repository of action datasets, each action dataset in the stored repository of action datasets having a corresponding set of command templates;

based on a determination that each of two or more action datasets identified in the search has at least one command template that corresponds to the command representation, accessing, by the digital assistant device, stored profile information associated with the digital assistant device;

determining, by the digital assistant device, that the accessed profile information includes a piece of device usage data that corresponds to a particular action dataset of the two or more action datasets; and

based on the determination that the accessed profile information includes the corresponding piece of device usage data, interpreting, by the digital assistant device, the particular action dataset to perform a set of emulated touch operations that corresponds to the command representation.

2. The computer-implemented method of claim 1, wherein the command representation includes a generated alphanumeric text representation of the received speech data.

3. The computer-implemented method of claim 1, further comprising:

communicating, by the digital assistant device, the generated query to a remote server device, wherein the remote server device is configured to search the stored repository of action datasets and determine that the included command representation corresponds to the associated set of command templates for each of the two or more action datasets.

4. The computer-implemented method of claim 1, wherein the piece of device usage data is determined to correspond to the particular action dataset based on a determined highest correlation between the piece of device usage data and the particular action dataset.

5. The computer-implemented method of claim 4, wherein the device usage data includes data corresponding to any one of a received command representation history, a determined relevant action dataset history, and metadata associated with the determined relevant action dataset history.

6. The computer-implemented method of claim 1, further comprising:

modifying, by the digital assistant device, the command representation to include the determined corresponding piece of device usage data.

7. The computer-implemented method of claim 6, wherein the modified command representation is employed to select the particular action dataset.

8. The computer-implemented method of claim 6, further comprising:

communicating, by the digital assistant device, the modified command representation to a remote server device, wherein the remote server device is configured to communicate the corresponding particular action dataset to the digital assistant device.

9. The computer-implemented method of claim 1, further comprising:

storing, by the digital assistant device, the obtained command representation and the corresponding particular action dataset as a command-action pair to the stored profile information.

10. A non-transitory computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the one or more processors to:

obtain a command representation that is generated based on speech data received as input via a digital assistant device;

identify, from a stored repository of action datasets, two or more action datasets that each includes an associated set of command templates determined to correspond to the obtained command representation;

access profile information associated with the digital assistant device based on the identification of the two or more action datasets; and

based on a determination that the accessed profile information includes a piece of device usage data that corresponds to a particular action dataset of the identified two or more action datasets, cause the digital assistant device to interpret a set of emulated touch operations included in the particular action dataset.

11. The media of claim 10, wherein the piece of device usage data is determined to correspond to the particular action dataset based on a determined highest correlation between the particular action dataset and other action datasets of the two or more action datasets.

12. The media of claim 10, wherein the device usage data includes data corresponding to at least one of a received command representation history, an interpreted action dataset history, and a history of application characteristics associated with action datasets interpreted by the digital assistant device.

13. The media of claim 10, wherein the piece of device usage data is determined to correspond to the particular action dataset based on a determined highest correlation between the piece of device usage data and the particular action dataset.

14. The media of claim 10, wherein the instructions further cause the one or more processors to:

modify the command representation to include the determined corresponding piece of device usage data.

15. The media of claim 14, wherein the modified command representation is employed to select the particular action dataset.

16. The media of claim 10, further comprising:

store the obtained command representation and the corresponding particular action dataset as a command-action pair to the stored profile information.

17. A computerized system for crowdsourcing digital assistant operations to digital assistant devices comprising:

one or more processors; and

one or more non-transitory computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the one or more processors to;

receive a command representation from a digital assistant device;

identify, from a repository of stored action datasets, two or more action datasets that each includes an associated set of command templates determined to correspond to the received command representation;

access a memory that includes profile information associated with the digital assistant device based on the identification of the two or more action datasets, the associated profile information including device usage data collected by the digital assistant device;

based on a determination that the profile information includes a piece of collected device usage data that corresponds to a particular action dataset of the identified two or more action datasets, cause the digital assistant device to interpret a set of operations included in the particular action dataset, wherein a set of touch operations is performed based on the interpreted set of operations.

18. The system of claim 17, wherein the piece of device usage data is determined to correspond to the particular action dataset based on a determined highest correlation between the piece of device usage data and the particular action dataset.

19. The system of claim 17, wherein the one or more non-transitory computer storage media storing computer-usable instructions that, when used by the one or more processors, further cause the one or more processors to:

modify the command representation to include the determined corresponding piece of device usage data, the modified command representation being employed to cause the digital assistant device to interpret the set of operations included in the particular action dataset.