Granted Patent B2

US 12,536,381 · App. 18/667,370 · Granted Jan 27, 2026

Inventor: Tal Haguel (Petah Tikva, IL)

Assignee: NICE LTD.

G06F40/30G06Q10/1093

Abstract

Claims (66)

1 . A stress detection system comprising:

a processor and a non-transitory computer readable medium operably coupled thereto, the non-transitory computer readable medium comprising a plurality of instructions stored in association therewith that are accessible to, and executable by, the processor, to perform operations which comprise:

building a library comprising previously identified stressful sentences and stressful phrases;

receiving, by a trained neural network model, the library;

receiving, by the trained neural network model, a text interaction between a customer and an agent;

calculating, by the trained neural network model, a cosine similarity score between each stressful sentence or stressful phrase in the library and each sentence in the text interaction;

determining, by the trained neural network model, a probability that the text interaction is stressful based on the calculated cosine similarity score;

determining that a percentage of stressful interactions for the agent in a time interval is greater than a threshold percentage;

providing a manager with recommended actions to decrease stress on the agent;

receiving, from the manager, a selection of one or more of the recommended actions; and

implementing the one or more recommended actions.

2 . The stress detection system of claim 1 , wherein building the library comprises:

receiving sentences and phrases in a customer interaction that show stress;

reducing the number of sentences and phrases by:

grouping the sentences and phrases into clusters via a hierarchical cluster algorithm,

calculating a cosine similarity score between each sentence and phrase in each cluster, and

identifying sentences and phrases in each cluster having a cosine similarity score above a threshold score; and

using the identified sentences and phrases to build the library.

3 . The stress detection system of claim 2 , wherein the operations further comprise identifying the sentences and phrases in the customer interaction that show stress by applying an audio stress detection algorithm to the sentences and phrases in the customer interaction or labeling predefined words in the sentences or phrases in the customer interaction.

4 . The stress detection system of claim 2 , wherein the hierarchical cluster algorithm comprises an agglomerative clustering algorithm.

5 . The stress detection system of claim 1 , wherein the time interval is a day.

6 . The stress detection system of claim 1 , wherein the recommended actions comprise updating a work shift, scheduling an intervention, scheduling a day off, or any combination thereof.

7 . The stress detection system of claim 1 , wherein the text interaction between the customer and the agent comprises text of one or more of a telephone call, a facsimile transmission, an e-mail, a chat, a web interaction, a voice over IP (“VOIP”), a video, or any combination thereof.

8 . The stress detection system of claim 1 , wherein the operations further comprise storing the library and the trained neural network model.

9 . A method for detecting and treating stress in a contact center, which comprises:

building a library comprising previously identified stressful sentences and stressful phrases;

receiving, by a trained neural network model, the library;

receiving, by the trained neural network model, a text interaction between a customer and an agent;

calculating, by the trained neural network model, a cosine similarity score between each stressful sentence or stressful phrase in the library and each sentence in the text interaction;

determining, by the trained neural network model, a probability that the text interaction is stressful based on the calculated cosine similarity score;

determining that a percentage of stressful interactions for the agent in a time interval is greater than a threshold percentage;

providing a manager with recommended actions to decrease stress on the agent;

receiving, from the manager, a selection of one or more of the recommended actions; and

implementing the one or more recommended actions.

10 . The method of claim 9 , wherein building the library comprises:

receiving sentences and phrases in a customer interaction that show stress;

reducing the number of sentences and phrases by:

grouping the sentences and phrases into clusters via a hierarchical cluster algorithm,

calculating a cosine similarity score between each sentence and phrase in each cluster, and

identifying sentences and phrases in each cluster having a cosine similarity score above a threshold score; and

using the identified sentences and phrases to build the library.

11 . The method of claim 10 , which further comprises identifying the sentences and phrases in the customer interaction that show stress by applying an audio stress detection algorithm to the sentences and phrases in the customer interaction or labeling predefined words in the sentences or phrases in the customer interaction.

12 . The method of claim 10 , wherein the hierarchical cluster algorithm comprises an agglomerative clustering algorithm.

13 . The method of claim 9 , wherein the recommended actions comprise updating a work shift, scheduling an intervention, scheduling a day off, or any combination thereof.

14 . The method of claim 9 , wherein the text interaction between the customer and the agent comprises text of one or more of a telephone call, a facsimile transmission, an e-mail, a chat, a web interaction, a voice over IP (“VoIP”), a video, or any combination thereof.

15 . The method of claim 9 , which further comprises storing the library and the trained neural network model.

16 . A non-transitory computer-readable medium having stored thereon computer-readable instructions executable by a processor to perform operations which comprise:

building a library comprising previously identified stressful sentences and stressful phrases;

receiving, by a trained neural network model, the library;

receiving, by the trained neural network model, a text interaction between a customer and an agent;

calculating, by the trained neural network model, a cosine similarity score between each stressful sentence or stressful phrase in the library and each sentence in the text interaction;

determining, by the trained neural network model, a probability that the text interaction is stressful based on the calculated cosine similarity score;

determining that a percentage of stressful interactions for the agent in a time interval is greater than a threshold percentage;

providing a manager with recommended actions to decrease stress on the agent;

receiving, from the manager, a selection of one or more of the recommended actions; and

implementing the one or more recommended actions.

17 . The non-transitory computer-readable medium of claim 16 , wherein building the library comprises:

receiving sentences and phrases in a customer interaction that show stress;

reducing the number of sentences and phrases by:

grouping the sentences and phrases into clusters via a hierarchical cluster algorithm,

calculating a cosine similarity score between each sentence and phrase in each cluster, and

identifying sentences and phrases in each cluster having a cosine similarity score above a threshold score; and

using the identified sentences and phrases to build the library.

18 . The non-transitory computer-readable medium of claim 17 , wherein the operations further comprise identifying the sentences and phrases in the customer interaction that show stress by applying an audio stress detection algorithm to the sentences and phrases in the customer interaction or labeling predefined words in the sentences or phrases in the customer interaction.

19 . The non-transitory computer-readable medium of claim 16 , wherein the recommended actions comprise updating a work shift, scheduling an intervention, scheduling a day off, or any combination thereof.

20 . The non-transitory computer-readable medium of claim 16 , wherein the text interaction between the customer and the agent comprises text of one or more of a telephone call, a facsimile transmission, an e-mail, a chat, a web interaction, a voice over IP (“VoIP”), a video, or any combination thereof.

Assignments (2)

Continuity (1)

References Cited (9)