Contact resolution for communications systems

• US 10,706,843 B1
• Filed: 03/09/2017
• Issued: 07/07/2020
• Est. Priority Date: 03/09/2017
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Elangovan, Someshwaran, Nandyal, Aparna, Kancharla, Venkatesh, Rajendran, Arun, Kshirsagar, Sumedha Arvind, Parker, Christopher Geiger
• Primary Examiner(s): Yen, Eric
• Application Number:
  US15/454,832
• Time in Patent Office: 1,216 Days
• CPC Class Codes: G06F 16/156 Query results presentation
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/0635 updating or merging of old and new templates; Mean values; Weighting
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/225 Feedback of the input speech
• CPC Class Codes: H04L 61/4594 Address books, i.e. directories containing contact information about correspondents telephone directories in user terminals H04M1/27453
• CPC Class Codes: H04M 1/271 controlled by voice recognition

Abstract

Methods and systems for performing contact resolution are described herein. When initiating a communications session using a voice activated electronic device, a contact name may be resolved to determine an appropriate contact with which the communications session may be directed to. Contacts from an individual'"'"'s contact list may be queried to determine a listing of probable contacts associated with the contact name, and contact identifiers associated with the contact may be determined. Using one or more rules for disambiguating between similar contact names, a single contact may be identified, and a communications session with that contact may be initiated.

Claims

1. A method, comprising:

receiving, at a computing system, first audio data representing a first utterance requesting that a first communications session be established with a contact, the first audio data being received from a first electronic device corresponding to a first account identifier;

generating first text data representing the first audio data;

determining that the first utterance corresponds to a communications intent, the communications intent including a contact name slot to be resolved;

determining, using the first text data, a first text value associated a portion of the first utterance corresponding to the contact, the first text value to be used for resolving an entity for the contact name slot;

accessing a contact list associated with the first account identifier;

determining that there is a first contact entry from the contact list that substantially matches the first text value;

determining that there is a second contact entry from the contact list that substantially matches the first text value;

determining a first contact identifier associated with the first contact entry;

determining a second contact identifier associated with the second contact entry;

determining that a contact identifier skip list includes the second contact identifier indicating that the second contact identifier is to be removed from consideration as a possible value to be associated with the contact name slot, the contact identifier skip list corresponding to at least one contact identifier associated with a corresponding contact entry from the contact list that has been determined to substantially match the first text value and indicated by a user as differing from the contact;

determining second text data representing a message asking for the first contact entry to be confirmed;

generating second audio data representing the second text data; and

sending the second audio data to the first electronic device.

2. The method of claim 1, further comprising:

receiving, at the computing system, and prior to the first audio data being received, third audio data representing a second utterance requesting that a second communications session be established with the contact, the third audio data being received from the first electronic device;

generating third text data representing the third audio data;

determining that the second utterance corresponds to the communications intent;

determining, using the third text data, that the first text value is associated with the portion;

accessing the contact list;

determining that the second contact entry substantially matches the first text value;

determining fourth text data representing a confirmation message asking whether the second contact entry is an intended contact for establishment of the second communications session;

generating fourth audio data representing the fourth text data;

sending the fourth audio data to the first electronic device;

receiving, at the computing system, fifth audio data representing a response utterance to the confirmation message, the fifth audio data being received from the first electronic device;

generating fifth text data representing the fifth audio data;

determining, using the fifth text data, that the second contact entry is different than the contact; and

adding the second contact identifier to the contact identifier skip list such that a subsequent request to establish a third communications session with the contact that identifies the second contact entry causes the second contact entry to be removed from consideration.

3. The method of claim 1, further comprising:

determining, prior to determining that the contact identifier skip list includes the second contact identifier, that there is a third contact entry from the contact list that substantially matches the first text value;

determining a third contact identifier associated with the third contact entry;

obtaining first contact information associated with the first contact identifier, the first contact information indicating a first group account on a communications system associated with the computing system that a first contact associated with the first contact entry is a member of;

obtaining second contact information associated with the third contact identifier, the second contact information indicating that a second contact associated with the third contact entry is unassociated with any group accounts on the communications system; and

determining that the first communications session to be established with the contact is only to be made to contacts that are members of a group account, such that the third contact identifier is removed from consideration because the second contact information indicates that the second contact is unassociated with any group accounts.

4. The method of claim 1, further comprising:

determining, prior to determining that the contact identifier skip list includes the second contact identifier, that there is a third contact entry from the contact list that substantially matches the first text value;

determining a third contact identifier associated with the third contact entry;

obtaining first contact information associated with the first contact identifier, the first contact information indicating a first group account on a communications system associated with the computing system that a first contact associated with the first contact entry is a member of;

obtaining second contact information associated with the third contact identifier, the second contact information indicating that that a third contact associated with the third contact entry is an additional member of the first group account;

determining that the first contact entry and the third contact entry are likely duplicate contact entries based on the first contact entry and the third contact entry both substantially matching the first text value and both the first contact entry and the third contact entry being members of the first group account; and

removing the third contact identifier from consideration as the entity.

5. A method, comprising:

receiving, from a first device associated with a first identifier, first audio data representing a first utterance;

determining that the first audio data corresponds to an intent to communicate with a contact;

determining a portion of the first audio data corresponds to first data to be used for resolving an entity;

determining a first contact list associated with the first identifier, the first contact list comprising at least a first entry represented by a first contact name value and a second entry represented by a second contact name value;

selecting the first entry based at least in part on the first data;

selecting the second entry based at least in part on the first data;

determining that a first skip list comprises the second entry, the first skip list comprising contact entries to be ignored;

determining first text data requesting confirmation of at least the first entry;

generating, in response to the first text data being determined, second audio data representing the first text data; and

sending the second audio data to the first device.

6. The method of claim 5, prior to the first audio data being received, further comprising:

receiving, a second request to communicate with the contact;

determining that the first contact list comprises the second entry;

determining second text data requesting confirmation of the second entry;

generating third audio data representing the second text data;

sending the third audio data to the first device;

receiving fourth audio data representing a second utterance comprising a response to the third audio data;

determining that the fourth audio data represents an indication that the second entry differs from the contact; and

determining the second entry is to be maintained on the first skip list.

7. The method of claim 5, further comprising:

receiving, from the first device, third audio data representing a second utterance comprising a response to the second audio data;

generating second text data representing the third audio data;

determining, using the second text data, that the first entry has been confirmed; and

causing a communication session to be initiated with a first communications identifier associated with the first entry.

8. The method of claim 5, further comprising:

determining, prior to determining that the first skip list comprises the second entry, that the first contact list comprises a third entry represented by a third contact name value;

selecting the third entry based at least in part on the first data;

receiving first contact information associated with the first entry, second contact information associated with the second entry, and third contact information associated with the third entry;

determining that the third entry is a duplicate of the first entry based, at least in part, on the first contact information and the third contact information both comprising a same communications identifier; and

determining that the third entry is to be ignored.

9. The method of claim 5, further comprising:

determining, prior to the first skip list being determined to comprise the second entry, a first confidence score indicating a first likelihood that the first contact name value comprises the first data;

determining a second confidence score indicating a second likelihood that the second contact name value comprises the second first data; and

determining that the first confidence score and the second confidence score both exceed a predefined confidence score threshold indicating that both the first contact name value and the second contact name value substantially match the first data.

10. The method of claim 5, further comprising:

receiving, from the first device, a second request to communicate with the contact;

determining that the first entry differs from the contact;

determining that a maximum number of attempts to resolve the contact have been performed; and

rendering an error prompt for the first device.

11. The method of claim 5, further comprising:

receiving third audio data representing a second utterance comprising a response to the second audio data;

generating second text data representing the third audio data;

determining, the second text data indicates that the first entry differs from the contact;

adding the first entry to the first skip list;

determining, based at least in part on the first skip list comprising the first entry and the second entry, third text data indicating that the contact is unable to be resolved;

generating fourth audio data representing the third text data; and

sending the fourth audio data to the first device.

12. The method of claim 5, further comprising:

determining, prior to the first skip list being determined to comprise the second entry, that the first contact list comprises a third entry represented by a third contact name value;

selecting the third entry based at least in part on the first data;

receiving first contact information associated with the first entry, second contact information associated with the second entry, and third contact information associated with the third entry;

determining, based at least in part on the third contact information, a number of communications with a communications identifier associated with the third entry;

causing, based at least in part on the number of communications being less than a predefined communication activity threshold, that the third entry is infrequently communicated with; and

adding the third entry to the first skip list.

13. A computing system, comprising:

at least one processor;

communications circuitry; and

memory including instructions operable to be executed by the at least one processor to configure the system to;

receive, from a first device associated with a first identifier, first audio data representing a first utterance;

determine that the first audio data corresponds to an intent to communicate with a contact;

determine a portion of the first audio data corresponds to a first data to be used for resolving an entity;

determine a first contact list associated with the first identifier, the first contact list comprising at least a first entry represented by a first contact name value and a second entry represented by a second contact name value;

select the first entry based at least in part on the first data;

select the second entry based at least in part on the first data;

determine that a first skip list comprises the second entry, the first skip list comprising entries to be ignored;

determine first text data requesting confirmation of at least the first entry;

generate, in response to the first text data being determined, second audio data representing the first text data; and

send the second audio data to the first device.

14. The computing system of claim 13, wherein the memory further comprises instructions that prior to the first audio data being received, further configure the system to:

receive, a second request to communicate with the contact;

determine that the first contact list comprises the second entry;

determine second text data requesting confirmation of the second entry;

generate third audio data representing the second text data;

send the third audio data to the first device;

receive fourth audio data representing a second utterance comprising a response to the third audio data;

determine the fourth audio data represents an indication that the second entry differs from the contact; and

determine the second entry is to be maintained on the first skip list.

15. The computing system of claim 13, wherein the at least one processor is further operable to:

determine, prior to the first skip list being determined to comprise the second entry, that the first contact list comprises a third entry represented by a third contact name value;

select the third entry based at least in part on the first data;

receive first contact information associated with the first entry, second contact information associated with the second entry, and third contact information associated with the third entry;

determine that the third entry is a duplicate of the first entry based, at least in part, on the first contact information and the third contact information both comprising a same communications identifier; and

determine that the third entry is to be ignored.

16. The computing system of claim 13, wherein the at least one processor is further operable to:

determine, prior to the first skip list being determined to comprise the second entry, a first confidence score indicating a first likelihood that the first contact name value comprises the first data;

determine a second confidence score indicating a second likelihood that the second contact name value comprises the first data; and

determine that the first confidence score and the second confidence score both exceed a predefined confidence score threshold indicating that both the first contact name value and the second contact name value substantially match the first data.

17. The computing system of claim 13, wherein the at least one processor is further operable to:

receive, from the first device, a second request to communicate with the contact;

determine that the first entry differs from the contact;

determine that a maximum number of attempts to resolve the contact have been performed; and

render an error prompt for the first device.

18. The computing system of claim 13, wherein the memory further comprises instructions that further configure the system to:

receive third audio data representing a second utterance comprising a response to the second audio data;

generate second text data representing the third audio data;

determine, the second text data indicates the first entry differs from the contact;

add the first entry to the first skip list;

determine, based at least in part on the first skip list comprising the first entry and the second entry, third text data indicating that the contact is unable to be resolved;

generate fourth audio data representing the third text data; and

send the fourth audio data to the first device.

19. The computing system of claim 13, wherein the at least one processor is further operable to:

determine, prior to the first skip list being determined to comprise the second entry, that the first contact list comprises a third entry represented by a third contact name value;

select the third entry based at least in part on the first data;

receive first contact information associated with the first entry, second contact information associated with the second entry, and third contact information associated with the third entry;

determine, based at least in part on the third contact information, a number of communications with a communications identifier associated with the third entry;

cause, based at least in part on the number of communications being less than a predefined communication activity threshold, that the third entry is infrequently communicated with; and

add the third entry to the first skip list.