Speech recognition system and method using group call statistics

• US 8,498,865 B1
• Filed: 02/17/2011
• Issued: 07/30/2013
• Est. Priority Date: 11/30/2004
• Status: Active Grant

• Current Assignee: Stryker Corporation
• Original Assignee: Vocera Communications, Inc. (Stryker Corporation)
• Inventors: Shostak, Robert E.
• Primary Examiner(s): Opsasnick, Michael N
• Application Number:
  US13/029,954
• Time in Patent Office: 894 Days
• Field of Search: 704/252, 704/276
• US Class Current: 704/252
• CPC Class Codes: G10L 15/04 Segmentation; Word boundary detection
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 2015/228 of application context

Abstract

An enhanced speech recognition system and method are provided that may be used with a voice recognition wireless communication system. The enhanced speech recognition system and method take advantage of group to group calling statistics to improve the recognition of names by the speech recognition system.

Claims

1. A computer implemented speech recognition system, comprising:

a communication system having one or more groups of users;

a speech recognition engine that is part of the communication system and generates a set of voice command interpretations based on a voice command of a user and a set of grammar, the voice command including a name of a person; and

the speech recognition engine further comprises a group unit that adjusts the set of grammar used by the speech recognition engine based on group to group calling statistics that modify a weighting of the set of grammar to facilitate the recognition of the name of a person used in the voice command, the group unit further comprising a call data gathering unit that gathers, for each call, a group to which a calling party is assigned and a group to which a called party is assigned to generate the group to group calling statistics for a particular group.

2. The system of claim 1, wherein the group unit further comprises a group probability unit that calculates, for a particular group, a probability that each other group in the communication system is called by users of the particular group based on the group to group calling statistics.

3. The system of claim 2, wherein the group unit further comprises a grammar modifier unit that modifies a grammar expression submitted to the speech recognition engine based on the probability that each other group in the communication system is called by users of the particular group.

4. The system of claim 1, wherein the call data gathering unit continuously gathers, for each call, a group to which a calling party is assigned and a group to which a called party is assigned to generate the group to group calling statistics for a particular group.

5. The system of claim 1, wherein the communication system further comprises a controlling computer, one or more concentrators connected to the controlling computer by a network, a badge that communicates using a protocol with one of the concentrators and wherein the controlling computer uses the speech recognition engine to receive a voice command from a particular user through the badge and interpret the voice command of the user to generate a set of voice commands interpretations.

6. The system of claim 1, wherein the communication system further comprises a controlling computer, an access device with a badge application that communicates with the controlling computer wherein the controlling computer uses the speech recognition engine to receive a voice command from a particular user through the badge application and interpret the voice command of the user to generate a set of voice commands interpretations.

7. The system of claim 1, wherein the one or more groups of users further comprises one or more departments in a health care industry.

8. A computer implemented speech recognition method in a communication system having one or more groups of users wherein the communication system has a speech recognition engine that generates a set of voice command interpretations based on a voice command of a user and grammar, the voice command including a name of a person, the method comprising:

adjusting, using a group unit of the speech recognition engine, the grammar used by the speech recognition engine based on group to group calling statistics, wherein adjusting the set of grammar further comprises gathering, using a call data gathering unit of the speech recognition engine, for each call, a group to which a calling party is assigned and a group to which a called party is assigned to generate the group to group calling statistics for a particular group; and

modifying, using the group unit of the speech recognition engine, a weighting of the grammar to facilitate the recognition of a name of a person used in the voice command.

9. The method of claim 8, wherein adjusting the set of grammar further comprises calculating, using a group probability unit of the speech recognition engine, for a particular group, a probability that each other group in the communication method is called by users of the particular group based on the group to group calling statistics.

10. The method of claim 8, wherein the gathering further comprises continuously gathering, for each call, a group to which a calling party is assigned and a group to which a called party is assigned to generate the group to group calling statistics for a particular group.

11. The method of claim 8, wherein the one or more groups of users further comprises one or more departments in a health care industry.

12. A computer implemented speech recognition system, comprising:

a controlling computer;

one or more concentrators connected to the controlling computer by a network;

a computing device that communicates using a protocol with one of the concentrators;

wherein the controlling computer further comprises a speech recognition engine that receives a voice command from a particular user through the computing device and interprets the voice command of the user to generate a set of voice commands interpretations based on a voice command of a user and grammar wherein the voice command includes a name of a person and wherein the system has one or more groups of user; and

the speech recognition engine further comprises a group unit that adjusts the set of grammar used by the speech recognition engine based on group to group calling statistics that modify a weighting of the grammar to facilitate the recognition of a name of a person used in the voice command, wherein the group unit further comprises a call data gathering unit that gathers, for each call, a group to which a calling party is assigned and a group to which a called party is assigned to generate the group to group calling statistics for a particular group.

13. The system of claim 12, wherein the group unit further comprises a group probability unit that calculates, for a particular group, a probability that each other group in the communication system is called by users of the particular group based on the group to group calling statistics.

14. The system of claim 13, wherein the group unit further comprises a grammar modifier unit that modifies the grammar expression submitted to the speech recognition engine based on the probability that each other group in the communication system is called by users of the particular group.

15. The system of claim 12, wherein the call data gathering unit continuously gathers, for each call, a group to which a calling party is assigned and a group to which a called party is assigned to generate the group to group calling statistics for a particular group.

16. The system of claim 12, wherein the computing device further comprises one or more a badge and a computing device with a badge application.

17. The system of claim 12, wherein the one or more groups of users further comprises one or more departments in a health care industry.