Voice and speech recognition for call center feedback and quality assurance

• US 10,404,859 B1
• Filed: 10/03/2018
• Issued: 09/03/2019
• Est. Priority Date: 06/29/2015
• Status: Active Grant

• Current Assignee: State Farm Mutual Automobile Insurance Company (State Farm)
• Original Assignee: State Farm Mutual Automobile Insurance Company (State Farm)
• Inventors: Hernandez, Sylvia
• Primary Examiner(s): Nguyen, Khai N.
• Application Number:
  US16/150,699
• Time in Patent Office: 335 Days
• Field of Search: 37921001, 37926501, 37926502, 37926503, 37926504, 37926505, 37926506, 37926507, 37926508, 37926509, 3792651, 37926511, 37926512, 37926513, 37926514, 37926601, 37926602, 37926603, 37926604, 37926605, 37926606, 37926607, 37926608, 37926609, 3792661
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/04 Training, enrolment or model building
• CPC Class Codes: G10L 17/08 Use of distortion metrics or a particular distance between probe pattern and reference templates
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/63 for estimating an emotional state
• CPC Class Codes: G10L 25/87 Detection of discrete points within a voice signal
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2201/41 using speaker recognition
• CPC Class Codes: H04M 2203/40 related to call centers
• CPC Class Codes: H04M 2203/401 Performance feedback
• CPC Class Codes: H04M 3/42 Systems providing special services or facilities to subscribers specially adapted for wireless communication networks H04W4/00
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements

Abstract

A computer-implemented method for providing an objective evaluation to a customer service representative regarding his performance during an interaction with a customer may include receiving a digitized data stream corresponding to a spoken conversation between a customer and a representative; converting the data stream to a text stream; generating a representative transcript that includes the words from the text stream that are spoken by the representative; comparing the representative transcript with a plurality of positive words and a plurality of negative words; and generating a score that varies according to the occurrence of each word spoken by the representative that matches one of the positive words, and/or the occurrence of each word spoken by the representative that matches one of the negative words. Tone of voice, as well as response time, during the interaction may also be monitored and analyzed to adjust the score, or generate a separate score.

Claims

1. A computer-implemented method for providing an objective evaluation to a customer service representative regarding his performance during an interaction with a customer, the computer-implemented method comprising:

comparing, via one or more processors, words which are spoken by the customer service representative during an oral conversation between the customer service representative and the customer with (i) positive words, and (ii) negative words; and

calculating and displaying on a display, via the one or more processors, a first score that varies according to (a) each word spoken by the customer service representative that matches one of the positive words, and (b) each word spoken by the customer service representative that matches one of the negative words to facilitate an objective evaluation of the customer interaction such that the first score is increased corresponding to usage of positive words by the customer service representative, and decreased corresponding to usage of negative words by the customer service representative.

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

receiving, via one or more processors, a digitized data stream corresponding to the oral conversation between the customer and the customer service representative; and

converting, via the one or more processors, the data stream to a text stream.

3. The computer-implemented method of claim 2, further comprising:

generating, via the one or more processors, a plurality of voice prints, each voice print derived from one of a plurality of periods of time during the conversation,

comparing, via the one or more processors, each voice print with a voice print of the customer service representative to determine an identity of the voice print,

indicating, via the one or more processors, a first set of voice prints that are associated with the customer service representative and a second set of voice prints that are associated with the customer,

matching, via the one or more processors, the first set of voice prints with words from the text stream that are spoken by the customer service representative, and

displaying, via the one or more processors, on the display a list of words from the customer service representative transcript that match one or more positive words, and a list of words from the customer service representative transcript that match one or more negative words.

4. The computer-implemented method of claim 2, further comprising:

analyzing, via the one or more processors, the data stream to determine a plurality of tone of voice values for the customer service representative, each tone of voice value derived from one of a plurality of periods of time during the conversation;

analyzing, via the one or more processors, the data stream to determine a plurality of tone of voice values for the customer, each tone of voice value derived from one of a plurality of periods of time during the conversation; and

generating and displaying on the display, via the one or more processors, a second score that varies according to whether the customer service representative changed his tone of voice in response to a change in tone of voice of the customer to facilitate the objective evaluation of a customer interaction.

5. The computer-implemented method of claim 4, further comprising determining, via the one or more processors, each occurrence when the customer'"'"'s tone of voice value is above a first threshold and determining, via the one or more processors, whether the tone of voice value for the customer service representative increases above a second threshold within a first time period after each occurrence.

6. The computer-implemented method of claim 5, further comprising

increasing, via the one or more processors, the second score for each occurrence when the tone of voice value for the customer service representative does not increase above the second threshold within the first time period;

decreasing, via the one or more processors, the second score for each occurrence when the tone of voice value for the customer service representative does increase above the second threshold within the first time period; and

displaying, via the one or more processors, on the display the second score and a list of occurrences when the customer service representative changed his tone of voice in response to a change in tone of voice of the customer.

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

identifying, via the one or more processors, a voice of the customer service representative and a voice of the customer;

determining, via the one or more processors, when the customer stops talking;

determining, via the one or more processors, when the customer service representative starts talking thereafter;

determining, via the one or more processors, a response time corresponding to a period of time that elapses between when the customer stops talking and the customer service representative starts talking;

generating and displaying, via the one or more processors, a third score that varies according to a value of the response time;

decreasing, via the one or more processors, the third score for every occurrence when the response time is less than a lower threshold or greater than an upper threshold; and

displaying, via the one or more processors, on the display a list of occurrences when the response time is less than a lower threshold or greater than an upper threshold.

8. A computing device for providing an objective evaluation to a customer service representative regarding his performance during an interaction with a customer, the computing device comprising:

a processing element in electronic communication with a memory element, the processing element configured to

compare words which are spoken by the customer service representative during an oral conversation between the customer service representative and the customer with (i) positive words, and (ii) negative words, and

calculate and display on a display a first score that varies according to (a) each word spoken by the customer service representative that matches one of the positive words, and (b) each word spoken by the customer service representative that matches one of the negative words to facilitate an objective evaluation of the customer interaction such that the first score is increased corresponding to usage of positive words by the customer service representative, and decreased corresponding to usage of negative words by the customer service representative.

9. The computing device of claim 8, wherein the processing element is further configured to receive a digitized data stream corresponding to the oral conversation between the customer and the customer service representative, and convert the data stream to a text stream.

10. The computing device of claim 9, wherein the processing element is further configured to

generate a plurality of voice prints, each voice print derived from one of a plurality of periods of time during the conversation,

compare each voice print with a voice print of the customer service representative to determine an identity of the voice print,

indicate a first set of voice prints that are associated with the customer service representative and a second set of voice prints that are associated with the customer,

match the first set of voice prints with words from the text stream that are spoken by the customer service representative, and

display on the display a list of words from the customer service representative transcript that match one or more positive words, and a list of words from the customer service representative transcript that match one or more negative words.

11. The computing device of claim 9, wherein the processing element is further configured to

analyze the data stream to determine a plurality of tone of voice values for the customer service representative, each tone of voice value derived from one of a plurality of periods of time during the conversation;

analyze the data stream to determine a plurality of tone of voice values for the customer, each tone of voice value derived from one of a plurality of periods of time during the conversation; and

generate and display on the display a second score that varies according to whether the customer service representative changed his tone of voice in response to a change in tone of voice of the customer to facilitate the objective evaluation of a customer interaction.

12. The computing device of claim 11, wherein the processing element is further configured to determine each occurrence when the customer'"'"'s tone of voice value is above a first threshold and determine whether the tone of voice value for the customer service representative increases above a second threshold within a first time period after each occurrence.

13. The computing device of claim 12, wherein the processing element is further configured to

increase the second score for each occurrence when the tone of voice value for the customer service representative does not increase above the second threshold within the first time period,

decrease the second score for each occurrence when the tone of voice value for the customer service representative does increase above the second threshold within the first time period, and

display on the display the second score and a list of occurrences when the customer service representative changed his tone of voice in response to a change in tone of voice of the customer.

14. The computing device of claim 8, wherein the processing element is further configured to

identify a voice of the customer service representative and a voice of the customer,

determine when the customer stops talking,

determine when the customer service representative starts talking thereafter,

determine a response time corresponding to a period of time that elapses between when the customer stops talking and the customer service representative starts talking,

generate and display a third score that varies according to a value of the response time,

decrease the third score for every occurrence when the response time is less than a lower threshold or greater than an upper threshold, and

display on the display a list of occurrences when the response time is less than a lower threshold or greater than an upper threshold.

15. A non-transitory computer-readable medium with an executable program stored thereon for providing an evaluation to a customer service representative regarding his performance during an interaction with a customer, wherein the program instructs a processing element of a computing device to perform the following:

comparing words which are spoken by the customer service representative during an oral conversation between the customer service representative and the customer with (i) positive words, and (ii) negative words; and

calculating and displaying on a display a first score that varies according to (a) each word spoken by the customer service representative that matches one of the positive words, and (b) each word spoken by the customer service representative that matches one of the negative words to facilitate an objective evaluation of the customer interaction such that the first score is increased corresponding to usage of positive words by the customer service representative, and decreased corresponding to usage of negative words by the customer service representative.

16. The non-transitory computer-readable medium of claim 15, wherein the program further instructs the processing element to receive a digitized data stream corresponding to the oral conversation between the customer and the customer service representative, and convert the data stream to a text stream.

17. The non-transitory computer-readable medium of claim 16, wherein the program further instructs the processing element to

generate a plurality of voice prints, each voice print derived from one of a plurality of periods of time during the conversation,

compare each voice print with a voice print of the customer service representative to determine an identity of the voice print,

indicate a first set of voice prints that are associated with the customer service representative and a second set of voice prints that are associated with the customer,

match the first set of voice prints with words from the text stream that are spoken by the customer service representative, and

display on the display a list of words from the customer service representative transcript that match one or more positive words, and a list of words from the customer service representative transcript that match one or more negative words.

18. The non-transitory computer-readable medium of claim 16,

wherein the program further instructs the processing element to analyze the data stream to determine a plurality of tone of voice values for the customer service representative, each tone of voice value derived from one of a plurality of periods of time during the conversation,

analyze the data stream to determine a plurality of tone of voice values for the customer, each tone of voice value derived from one of a plurality of periods of time during the conversation,

generate and display on the display a second score that varies according to whether the customer service representative changed his tone of voice in response to a change in tone of voice of the customer to facilitate the objective evaluation of a customer interaction, and

determine each occurrence when the customer'"'"'s tone of voice value is above a first threshold and determine whether the tone of voice value for the customer service representative increases above a second threshold within a first time period after each occurrence.

19. The non-transitory computer-readable medium of claim 18,

wherein the program further instructs the processing element to increase the second score for each occurrence when the tone of voice value for the customer service representative does not increase above the second threshold within the first time period,

decrease the second score for each occurrence when the tone of voice value for the customer service representative does increase above the second threshold within the first time period, and

display on the display the second score and a list of occurrences when the customer service representative changed his tone of voice in response to a change in tone of voice of the customer.

20. The non-transitory computer-readable medium of claim 15,

wherein the program further instructs the processing element to

identify a voice of the customer service representative and a voice of the customer,

determine when the customer stops talking,

determine when the customer service representative starts talking thereafter,

determine a response time corresponding to a period of time that elapses between when the customer stops talking and the customer service representative starts talking,

generate and display a third score that varies according to a value of the response time,

decrease the third score for every occurrence when the response time is less than a lower threshold or greater than an upper threshold, and

display on the display a list of occurrences when the response time is less than a lower threshold or greater than an upper threshold.