CN111881272A - Real-time early warning system and method for telephone service - Google Patents
Real-time early warning system and method for telephone service Download PDFInfo
- Publication number
- CN111881272A CN111881272A CN202010656209.9A CN202010656209A CN111881272A CN 111881272 A CN111881272 A CN 111881272A CN 202010656209 A CN202010656209 A CN 202010656209A CN 111881272 A CN111881272 A CN 111881272A
- Authority
- CN
- China
- Prior art keywords
- early warning
- module
- analysis
- real
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/182—Level alarms, e.g. alarms responsive to variables exceeding a threshold
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Abstract
The invention provides a real-time early warning system and an early warning method for telephone service, wherein the real-time early warning system comprises: the early warning system comprises an original data processing module, a voice analysis module, a model analysis module, an early warning filtering module, an early warning context extraction module, an early warning global management module and an early warning pushing module; the technical scheme provided by the invention can detect abnormal service problems such as abusive clients, slow response of the agents, abnormal emotions, client complaints and the like at the first time in the process of providing services for the clients by the agents, and can realize first-time discovery, first-time follow-up and first-time solution.
Description
Technical Field
The invention belongs to the technical field of telephone service quality evaluation, and particularly relates to a real-time early warning system and method for telephone service.
Background
Customer satisfaction with an agent service is an important way to assess and improve customer service experience in an enterprise call center business. As speech analysis continues to be applied in call centers, enterprise management layers place greater demands on the timeliness of the analysis, requiring abnormal service issues, such as abusive customers, slow response of the agents, abnormal emotions, customer complaints, etc., to be detected the first time an agent is servicing a customer. The method comprises the steps of finding in the first time, following in the first time and solving in the first time.
However, in the prior art, no effective solution is provided for real-time early warning of telephone services.
Disclosure of Invention
In order to overcome the lack of real-time early warning on telephone service in the prior art, the patent application provides a real-time early warning system and a real-time early warning method for telephone service, which can detect abnormal service problems such as abuse customers, slow response of the agents, abnormal emotions, customer complaints and the like at the first time in the process of providing services for customers by the agents. The method comprises the steps of finding in the first time, following in the first time and solving in the first time.
The technical scheme of the invention is as follows:
a real-time alerting system for telephony services, the real-time alerting system comprising: the system comprises an original data processing module, a voice analysis module, a model analysis module, an early warning filtering module, an early warning context extraction module, an early warning global management module and an early warning pushing module:
the original data processing module comprises a voice recognition module and a text data extraction module, wherein the voice recognition module is used for converting voice recognition of voice type data into a recognition text and voice attribute information; the text data extraction module is used for directly extracting the text data into an identification text;
the voice analysis module comprises a mute analysis module, a call interruption analysis module, a speech speed analysis module and an emotion analysis module, and is used for performing mute analysis, call interruption analysis, speech speed analysis and emotion analysis on the voice attribute information and obtaining an analysis conclusion;
the model analysis module executes model analysis on the recognition text converted after the voice recognition of the voice recognition module and the recognition text directly extracted by the text data extraction module through a pre-established early warning model, and judges whether relevant early warning occurs or not according to a model matching result;
the early warning filtering module is used for filtering the analysis results of the silence analysis module, the inserting speech analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module;
the early warning context extraction module is responsible for extracting context information when early warning occurs;
the early warning global management module performs centralized management on all seat early warning data in the whole real-time early warning system, and provides uniform early warning query support and early warning push support;
the early warning pushing module is responsible for pushing early warning information to user groups at all levels.
A real-time early warning method for telephone service, the real-time early warning method uses the real-time early warning system of the telephone service, its step is as follows:
the voice recognition module converts voice recognition of the voice type data into recognition text and voice attribute information; the text data extraction module directly extracts the text data into an identification text;
the silence analysis module, the inserting speech analysis module, the speech speed analysis module and the emotion analysis module execute silence analysis, inserting speech analysis, speech speed analysis and emotion analysis on the voice attribute information and obtain an analysis conclusion;
the model analysis module executes model analysis on the recognition text converted after the voice recognition of the voice recognition module and the recognition text directly extracted by the text data extraction module through a pre-established early warning model, and judges whether relevant early warning occurs or not according to a model matching result;
the early warning filtering module is used for filtering the analysis results of the silence analysis module, the inserting speech analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module;
the early warning context extraction module is responsible for extracting context information when early warning occurs;
the early warning global management module performs centralized management on all seat early warning data in the whole real-time early warning system, and provides uniform early warning query support and early warning push support;
the early warning pushing module is responsible for pushing early warning information to user groups at all levels.
Preferably, the analysis process of the early warning filtering module judges whether to early warn abnormal silence, whether to early warn abnormal robbery and insertion, whether to early warn abnormal speech rate, whether to early warn abnormal emotion, and whether to early warn abnormal text in real time according to the real-time early warning configuration.
Preferably, the context information includes context voice attribute information and a context identification text, and is used for directly knowing the context background of the occurrence of the early warning through the early warning information to extract the conversation context information including speaking time, speaking role and speaking content when the early warning occurs.
Preferably, the early warning pushing module needs to push according to a hierarchy of a call center service group multi-level architecture, and this pushing needs to satisfy: the parent user group can receive the early warning information of the child user group.
Preferably, when the early warning pushing module executes the early warning pushing, the early warning pushing module acquires the seat information of the service according to the call information when the early warning occurs, acquires the user group to which the service belongs according to the seat information, acquires the parent user group according to the user group, sequentially acquires the parent user group step by step until reaching the root user group, and simultaneously pushes the corresponding early warning information to the user groups at all levels.
Compared with the prior art, the invention has the advantages that:
1. the telephone service real-time early warning system and the telephone service real-time early warning method provide a unified real-time early warning scheme aiming at voice data and text data.
2. The silence analysis module, the inserting speech analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module are parallel early warning analysis technologies, detection problems of different early warning types can be solved, and a person skilled in the art can perform synchronous and asynchronous combination on the analysis modules to flexibly meet various early warning requirements.
3. The early warning pushing module provides adaptive support for a call center agent multi-level management mode, and can ensure that early warning events of high-level management and medium-level management can be supported.
4. The context information extraction comprises extracting context voice attribute information of the occurrence of the early warning call and extracting context recognition text, and is used for directly knowing the context background of the occurrence of the early warning through the early warning information.
5. The telephone service real-time early warning system and the telephone service real-time early warning method can detect abnormal service problems such as abuse customers, slow response of the agents, abnormal emotions, customer complaints and the like at the first time in the process of providing services for the customers by the agents, and achieve the purposes of first time discovery, first time follow-up and first time solution.
Drawings
Fig. 1 is a structural diagram of a real-time warning system for telephone service according to the present invention.
Fig. 2 is a flow chart illustrating steps of a real-time warning method for a telephone service according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
A real-time warning system for telephony services, the warning system comprising: the system comprises an original data processing module, a voice analysis module, a model analysis module, an early warning filtering module, an early warning context extraction module, an early warning global management module and an early warning pushing module:
1) the original data processing module comprises a voice recognition module and a text data extraction module, wherein voice type data is converted into recognition text and voice attribute information after being subjected to voice recognition by the voice recognition module; the text data is directly extracted into the recognition text through a text data extraction module.
2) The voice analysis module comprises a mute analysis module, a call interruption analysis module, a speech speed analysis module and an emotion analysis module and is used for performing mute analysis, call interruption analysis, speech speed analysis and emotion analysis on the voice attribute information and obtaining an analysis conclusion.
3) And the model analysis module executes model analysis on the recognition text converted after the voice recognition of the voice recognition module and the recognition text directly extracted by the text data extraction module through a pre-established early warning model, and judges whether relevant early warning occurs or not according to a model matching result.
4) The early warning filter module is used for filtering the analysis results of the silence analysis module, the call-inserting analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module, and the early warning analysis process is configured according to real-time early warning, such as whether abnormal silence is early warned, whether abnormal call-inserting is early warned, whether abnormal speech speed is early warned, whether abnormal emotion is early warned, whether abnormal text is early warned and the like.
5) The early warning context extraction module is responsible for extracting context information when early warning occurs, including context voice attribute information and context recognition text, and is used for directly knowing the context background when the early warning occurs through the early warning information and extracting conversation context information when the early warning occurs, including speaking time, speaking role and speaking content. The real-time early warning result comes from the real-time analysis result of the computer, and when the business falls to the ground, the customer may need to manually review. The early warning context information is key information during rechecking, and comprises conversation context information including speaking time, speaking role, speaking content and the like during early warning.
6) And the early warning global management module is used for carrying out centralized management on all seat early warning data in the whole real-time early warning system and providing unified early warning query support and early warning push support.
7) The early warning pushing module is responsible for pushing early warning information to user groups at all levels, and early warning pushing needs to be carried out according to the levels of the multi-level architecture of the call center service group. For example: the credit card group comprises an installment service group and an opening service group. And pushing according to the hierarchy of the service group when early warning is pushed. This pushing needs to satisfy: the parent user group can receive the early warning information of the child user group.
When early warning pushing is executed, seat information of service is acquired according to call information when early warning occurs, a user group to which the service belongs is acquired according to the seat information, a parent user group is acquired according to the user group, and the parent user group is sequentially acquired step by step until a root user group is reached. And then, the corresponding early warning information is simultaneously pushed to the user groups at all levels so as to meet the service requirement of early warning pushing.
A real-time early warning method for telephone service, the real-time early warning method uses the real-time early warning system of the telephone service, its step is as follows:
1) the voice recognition module converts voice recognition of the voice type data into recognition text and voice attribute information; the text data extraction module directly extracts the text data into an identification text;
2) the silence analysis module, the inserting speech analysis module, the speech speed analysis module and the emotion analysis module execute silence analysis, inserting speech analysis, speech speed analysis and emotion analysis on the voice attribute information and obtain an analysis conclusion;
3) the model analysis module executes model analysis on the recognition text converted after the voice recognition of the voice recognition module and the recognition text directly extracted by the text data extraction module through a pre-established early warning model, and judges whether relevant early warning occurs or not according to a model matching result;
4) the early warning filtering module is used for filtering the analysis results of the silence analysis module, the inserting speech analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module;
5) the early warning context extraction module is responsible for extracting context information when early warning occurs;
6) the early warning global management module performs centralized management on all seat early warning data in the whole real-time early warning system, and provides uniform early warning query support and early warning push support;
7) the early warning pushing module is responsible for pushing early warning information to user groups at all levels.
The above-described embodiments are merely preferred embodiments of the present invention, which is not intended to limit the present invention in any way. Those skilled in the art can make many changes, modifications, and equivalents to the embodiments of the invention without departing from the scope of the invention as set forth in the claims below. Therefore, equivalent variations made according to the idea of the present invention should be covered within the protection scope of the present invention without departing from the contents of the technical solution of the present invention.
Claims (10)
1. A real-time warning system for telephony services, comprising: the real-time early warning system comprises: the system comprises an original data processing module, a voice analysis module, a model analysis module, an early warning filtering module, an early warning context extraction module, an early warning global management module and an early warning pushing module:
the original data processing module comprises a voice recognition module and a text data extraction module, wherein the voice recognition module is used for converting voice recognition of voice type data into a recognition text and voice attribute information; the text data extraction module is used for directly extracting the text data into an identification text;
the voice analysis module comprises a mute analysis module, a call interruption analysis module, a speech speed analysis module and an emotion analysis module, and is used for performing mute analysis, call interruption analysis, speech speed analysis and emotion analysis on the voice attribute information and obtaining an analysis conclusion;
the model analysis module executes model analysis on the recognition text converted after the voice recognition of the voice recognition module and the recognition text directly extracted by the text data extraction module through a pre-established early warning model, and judges whether relevant early warning occurs or not according to a model matching result;
the early warning filtering module is used for filtering the analysis results of the silence analysis module, the inserting speech analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module;
the early warning context extraction module is responsible for extracting context information when early warning occurs;
the early warning global management module performs centralized management on all seat early warning data in the whole real-time early warning system, and provides uniform early warning query support and early warning push support;
the early warning pushing module is responsible for pushing early warning information to user groups at all levels.
2. The real-time pre-warning system for telephone services of claim 1, wherein: the analysis process of the early warning filtering module judges whether to early warn abnormal silence, whether to early warn abnormal robbery insertion, whether to early warn abnormal speech rate, whether to early warn abnormal emotion and whether to early warn abnormal text in real time according to real-time early warning configuration.
3. A real-time alerting system for telephone services of any one of claims 1-2, wherein: the context information comprises context voice attribute information and a context recognition text, and is used for directly knowing the context background of the early warning occurrence through the early warning information and extracting the conversation context information when the early warning occurs, wherein the conversation context information comprises speaking time, speaking role and speaking content.
4. A real-time alerting system for telephone services of any one of claims 1-3, wherein: the early warning pushing module needs to push according to the hierarchy of the call center service group multilevel architecture, and the pushing needs to meet the following requirements: the parent user group can receive the early warning information of the child user group.
5. A real-time alerting system of telephone services of any one of claims 1-4, wherein: when the early warning pushing module executes early warning pushing, the early warning pushing module acquires service seat information according to call information when early warning occurs, acquires a user group to which the service seat information belongs according to the seat information, acquires a parent user group according to the user group, sequentially acquires the parent user group step by step until reaching a root user group, and simultaneously pushes corresponding early warning information to the user groups at all levels.
6. A real-time early warning method for telephone service is characterized in that: the real-time early warning method using the real-time early warning system for telephone service according to any one of claims 1 to 5, comprising the steps of:
the voice recognition module converts voice recognition of the voice type data into recognition text and voice attribute information; the text data extraction module directly extracts the text data into an identification text;
the silence analysis module, the inserting speech analysis module, the speech speed analysis module and the emotion analysis module execute silence analysis, inserting speech analysis, speech speed analysis and emotion analysis on the voice attribute information and obtain an analysis conclusion;
the model analysis module executes model analysis on the recognition text converted after the voice recognition of the voice recognition module and the recognition text directly extracted by the text data extraction module through a pre-established early warning model, and judges whether relevant early warning occurs or not according to a model matching result;
the early warning filtering module is used for filtering the analysis results of the silence analysis module, the inserting speech analysis module, the speech speed analysis module, the emotion analysis module and the model analysis module;
the early warning context extraction module is responsible for extracting context information when early warning occurs;
the early warning global management module performs centralized management on all seat early warning data in the whole real-time early warning system, and provides uniform early warning query support and early warning push support;
the early warning pushing module is responsible for pushing early warning information to user groups at all levels.
7. The real-time pre-warning method for telephone service according to claim 6, wherein: the analysis process of the early warning filtering module judges whether to early warn abnormal silence, whether to early warn abnormal robbery insertion, whether to early warn abnormal speech rate, whether to early warn abnormal emotion and whether to early warn abnormal text in real time according to real-time early warning configuration.
8. A real-time alerting method of a telephone service of any one of claims 6-7, wherein: the context information comprises context voice attribute information and a context recognition text, and is used for directly knowing the context background of the early warning occurrence through the early warning information and extracting the conversation context information when the early warning occurs, wherein the conversation context information comprises speaking time, speaking role and speaking content.
9. A real-time alerting method of a telephone service of any one of claims 6-8, wherein: the early warning pushing module needs to push according to the hierarchy of the call center service group multilevel architecture, and the pushing needs to meet the following requirements: the parent user group can receive the early warning information of the child user group.
10. A real-time alerting method of a telephone service of any one of claims 6-9, wherein: when the early warning pushing module executes early warning pushing, the early warning pushing module acquires service seat information according to call information when early warning occurs, acquires a user group to which the service seat information belongs according to the seat information, acquires a parent user group according to the user group, sequentially acquires the parent user group step by step until reaching a root user group, and simultaneously pushes corresponding early warning information to the user groups at all levels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010656209.9A CN111881272A (en) | 2020-07-09 | 2020-07-09 | Real-time early warning system and method for telephone service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010656209.9A CN111881272A (en) | 2020-07-09 | 2020-07-09 | Real-time early warning system and method for telephone service |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111881272A true CN111881272A (en) | 2020-11-03 |
Family
ID=73150522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010656209.9A Pending CN111881272A (en) | 2020-07-09 | 2020-07-09 | Real-time early warning system and method for telephone service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111881272A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112511698A (en) * | 2020-12-03 | 2021-03-16 | 普强时代(珠海横琴)信息技术有限公司 | Real-time call analysis method based on universal boundary detection |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015131652A1 (en) * | 2014-09-19 | 2015-09-11 | 中兴通讯股份有限公司 | Method and device for processing warning information data |
| CN104951433A (en) * | 2015-06-24 | 2015-09-30 | 北京京东尚科信息技术有限公司 | Method and system for intention recognition based on context |
| CN209947074U (en) * | 2019-04-08 | 2020-01-14 | 张春强 | Fire control passageway prevents blockking up early warning system |
| CN111128241A (en) * | 2019-12-30 | 2020-05-08 | 上海浩琨信息科技有限公司 | Intelligent quality inspection method and system for voice call |
-
2020
- 2020-07-09 CN CN202010656209.9A patent/CN111881272A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015131652A1 (en) * | 2014-09-19 | 2015-09-11 | 中兴通讯股份有限公司 | Method and device for processing warning information data |
| CN104951433A (en) * | 2015-06-24 | 2015-09-30 | 北京京东尚科信息技术有限公司 | Method and system for intention recognition based on context |
| CN209947074U (en) * | 2019-04-08 | 2020-01-14 | 张春强 | Fire control passageway prevents blockking up early warning system |
| CN111128241A (en) * | 2019-12-30 | 2020-05-08 | 上海浩琨信息科技有限公司 | Intelligent quality inspection method and system for voice call |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112511698A (en) * | 2020-12-03 | 2021-03-16 | 普强时代(珠海横琴)信息技术有限公司 | Real-time call analysis method based on universal boundary detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108763499B (en) | Call quality inspection method, device, equipment and storage medium based on intelligent voice | |
| US11714793B2 (en) | Systems and methods for providing searchable customer call indexes | |
| US10410636B2 (en) | Methods and system for reducing false positive voice print matching | |
| US8145562B2 (en) | Apparatus and method for fraud prevention | |
| US8903859B2 (en) | Systems, methods, and media for generating hierarchical fused risk scores | |
| US8306814B2 (en) | Method for speaker source classification | |
| US9571652B1 (en) | Enhanced diarization systems, media and methods of use | |
| US8290951B1 (en) | Unstructured data integration with a data warehouse | |
| CA2564760C (en) | Speech analysis using statistical learning | |
| CN110176252A (en) | Intelligent sound quality detecting method and system based on risk management and control mode | |
| CN112511696A (en) | System and method for identifying bad content of call center AI engine | |
| CN110705926A (en) | Method, device and system for acquiring logistics object distribution information | |
| CN110677540A (en) | Intelligent voice recognition management system for consultation telephone of medical institution | |
| CN111831875A (en) | Data processing method, device, equipment and storage medium | |
| CN111881272A (en) | Real-time early warning system and method for telephone service | |
| CN107645613A (en) | The method and apparatus of service diverting search | |
| CN116501844A (en) | Voice keyword retrieval method and system | |
| WO2019140242A1 (en) | User support system with automatic message categorization | |
| CN115396549A (en) | Method for processing illegal call service device and electronic apparatus | |
| Lee et al. | A study on natural language call routing | |
| CN110728145A (en) | Method for establishing natural language understanding model based on recording conversation | |
| CN110784603A (en) | Intelligent voice analysis method and system for offline quality inspection | |
| CN110166640B (en) | System and method for establishing call center general information model | |
| CN114581103A (en) | Anti-telecommunication fraud service end user interaction system | |
| Roshan et al. | Capturing important information from an audio conversation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |