SYSTEM AND METHOD FOR DYNAMIC ASR BASED ON SOCIAL MEDIA

• US 20170186419A1
• Filed: 03/13/2017
• Published: 06/29/2017
• Est. Priority Date: 09/15/2012
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Avaya Incorporated
• Inventors: Erhart, George W., Matula, Valentine C., Skiba, David J.
• Granted Patent: US 10,134,391 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/685 using automatically derived transcript of audio data, e.g. lyrics speech recognition G10L15/00
• CPC Class Codes: G06Q 30/0281 Customer communication at a business location, e.g. providing product or service information, consulting
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/0635 updating or merging of old and new templates; Mean values; Weighting
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: H04L 51/52 for supporting social networking services
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/40 related to call centers
• CPC Class Codes: H04M 2203/655 Combination of telephone service and social networking messaging within social networks H04L51/52
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends
• CPC Class Codes: H04M 3/5191 interacting with the Internet

Abstract

System and method to adjust an automatic speech recognition (ASR) engine, the method including: receiving social network information from a social network; data mining the social network information to extract one or more characteristics; inferring a trend from the extracted one or more characteristics; and adjusting the ASR engine based upon the inferred trend. Embodiments of the method may further include: receiving a speech signal from a user; and recognizing the speech signal by use of the adjusted ASR engine. Further embodiments of the method may include: producing a list of candidate matching words; and ranking the list of candidate matching words by use of the inferred trend.

Claims

1. A method to adjust a lexicon of a contact center system, the method comprising:

receiving, by a social media gateway of a contact center, a plurality of messages from at least one social network, wherein each message comprises one or more message fields;

grouping, by the social media gateway, based on a content of the respective one more message fields, two or more of the messages to form a message grouping;

data mining, by a dialog engine of the contact center, the message grouping to extract one or more characteristics;

inferring, by the dialog engine, based on the one or more extracted characteristics, a trend between the messages comprising the message grouping; and

adding, by the dialog engine, one or more words or phrases related to the trend to the lexicon.

2. The method of claim 1, further comprising:

calculating, by the dialog engine, a magnitude of adjustment to weights of the added one or more words or phrases based upon a shaped sliding window.

3. The method of claim 2, further comprising:

adjusting, by the dialog engine, a speech recognition weighting of an ASR engine of the contact center based upon the calculated magnitude of adjustment.

4. The method of claim 3, further comprising:

receiving, by the ASR engine, a speech signal from a user; and

matching, by the ASR engine, the speech signal to words in the lexicon.

5. The method of claim 4, wherein matching the speech signal to words in the lexicon comprises:

producing a list of candidate matching words; and

ranking the list of candidate matching words.

6. The method of claim 3, further comprising:

estimating, by the dialog engine, a persistence of the trend; and

limiting, by the dialog engine, a duration of the adjustment to the speech recognition weighting based on the estimated persistence.

7. The method of claim 1, further comprising:

generating, by the dialog engine, a response to one or more of the plurality of messages.

8. The method of claim 7, wherein generating the response comprises one or more of:

creating one or more dialog data structures;

or

retrieving one or more existing dialog data structures.

9. The method of claim 1, further comprising:

searching, by the social media gateway, the at least one social network for one or more of additional messages or user context information.

10. The method of claim 1, wherein the message grouping comprises messages from two or more different social networks.

18. The system of claim 1, wherein the processor, when executing the executable instructions:

searches the at least one social network for one or more of additional messages or user context information.

11. A system of a contact center to adjust a lexicon of the contact center, the system comprising:

a server comprising;

a computer-readable storage medium, storing executable instructions; and

a processor coupled to the computer-readable storage medium, the processor, when executing the executable instructions;

receives a plurality of messages from at least one social network, wherein each message comprises one or more message fields;

based on a content of the respective one more message fields, groups two or more of the messages to form a message grouping;

data mines the message grouping to extract one or more characteristics;

based on the one or more extracted characteristics, infers a trend between the messages comprising the message grouping; and

adds one or more words or phrases related to the trend to the lexicon.

12. The system of claim 11, wherein the processor, when executing the executable instructions:

calculates a magnitude of adjustment to weights of the added one or more words or phrases based upon a shaped sliding window.

13. The system of claim 12, wherein the processor, when executing the executable instructions:

adjusts a speech recognition weighting of an ASR engine of the contact center based upon the calculated magnitude of adjustment.

14. The system of claim 13, wherein the processor, when executing the executable instructions:

receives a speech signal from a user; and

matches the speech signal to words in the lexicon, wherein matching the speech signal comprises;

producing a list of candidate matching words; and

ranking the list of candidate matching words.

15. The system of claim 13, wherein the processor, when executing the executable instructions:

estimates a persistence of the trend; and

limits a duration of the adjustment to the speech recognition weighting based on the estimated persistence.

16. The system of claim 11, wherein the processor, when executing the executable instructions:

generates a response to one or more of the plurality of messages.

17. The system of claim 16, wherein generating the response comprises one or more of:

creating one or more dialog data structures;

or

retrieving one or more existing dialog data structures.

19. The system of claim 11, wherein the message grouping comprises messages from two or more different social networks.

20. A method for adjusting a lexicon of a contact center system, the method comprising:

receiving, by a social media gateway of a contact center, social media information from at least one social network;

data mining, by a dialog engine of the contact center, the social media information to extract one or more characteristics;

inferring, by the dialog engine, a trend based on the one or more extracted characteristics; and

modifying, by the dialog engine, the lexicon based on the trend, wherein modifying the lexicon comprises at least one of;

adding one or more words or phrases related to the trend to the lexicon;

or

deleting one or more words or phrases previously added to the lexicon based on a previous trend.