CN117998014A - Intelligent IVR implementation method, terminal equipment and storage medium - Google Patents

Intelligent IVR implementation method, terminal equipment and storage medium Download PDF

Info

Publication number
CN117998014A
CN117998014A CN202410195587.XA CN202410195587A CN117998014A CN 117998014 A CN117998014 A CN 117998014A CN 202410195587 A CN202410195587 A CN 202410195587A CN 117998014 A CN117998014 A CN 117998014A
Authority
CN
China
Prior art keywords
matched
response mode
user
configuring
intention
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
Application number
CN202410195587.XA
Other languages
Chinese (zh)
Inventor
郑陈挺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xiamen Yeastar Digital Technology Co ltd
Original Assignee
Xiamen Yeastar Digital Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xiamen Yeastar Digital Technology Co ltd filed Critical Xiamen Yeastar Digital Technology Co ltd
Priority to CN202410195587.XA priority Critical patent/CN117998014A/en
Publication of CN117998014A publication Critical patent/CN117998014A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/527—Centralised call answering arrangements not requiring operator intervention
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/18—Speech classification or search using natural language modelling
    • G10L15/1822—Parsing for meaning understanding
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Telephonic Communication Services (AREA)

Abstract

The invention relates to an intelligent IVR realization method, terminal equipment and storage medium, wherein the method comprises the following steps: classifying intentions of incoming calls, and configuring a response mode corresponding to each intention; the configuration items in each response mode include at least intent, intent description, and action; when receiving an incoming call, sending the content of the incoming call and each configured response mode to a large language model together to obtain a matching result of the incoming call and each response mode, and if the matching result is matched, responding according to actions in the matched response mode; if the user intention prompt is not matched, playing the converted prompt which is not understood and identified in advance. The invention has the advantages of simple configuration, good user experience and high efficiency.

Description

Intelligent IVR implementation method, terminal equipment and storage medium
Technical Field
The present invention relates to the field of network telephony, and in particular, to an intelligent IVR implementation method, a terminal device, and a storage medium.
Background
The implementation method of the traditional non-intelligent IVR (automatic voice response service) comprises the following steps: (1) setting the number of the IVR on the IPPBX; (2) uploading IVR playing prompt tones; (3) setting an event of the 0-9 key. When the user calls into the IVR through the external line, the user first hears the play prompt of the IVR (e.g., XXX please press 1, XXX please press 2.) and then seeks the corresponding service through the key, as shown in fig. 1. This solution has the following problems:
(1) The configuration is troublesome: a person needs to be found to record a section of system prompt tone and upload the system prompt tone, and recording contents are in one-to-one correspondence with the set key events; the modification is also troublesome and the whole section of the prompt tone needs to be re-recorded. In addition, the key supports only 0-9, namely 10 response services, and if the response service exceeds the response service, a plurality of IVRs are required to be split and created for implementation.
(2) The user experience is poor: from entering the IVR, a large section of alert tones are heard before knowing which key is pressed to find the corresponding service.
In addition, the realization of the ordinary intelligent IVR only replaces a key part, the traditional AI model is adopted, the intention of the user can be simply identified, but after the intention is identified, a plurality of parameters of the user intention mapping action are needed to be obtained through a plurality of times of conversations with the user, for example, when the user calls in the IVR, the user is firstly prompted to guide what is wanted to be done, if the user is identified to be wanted to be the customer service manager, the user is further queried for 'please ask for which customer service manager is wanted to be found, please complain about his name', the user answers 'XXX', the answer content is the name, and if the answer 'I want to find the XXX customer service manager', the name cannot be intelligently resolved.
Disclosure of Invention
In order to solve the problems, the invention provides an intelligent IVR implementation method, terminal equipment and a storage medium.
The specific scheme is as follows:
An intelligent IVR implementation method, comprising: classifying intentions of incoming calls, and configuring a response mode corresponding to each intention; the configuration items in each response mode include at least intent, intent description, and action; when receiving an incoming call, sending the content of the incoming call and each configured response mode to a large language model together to obtain a matching result of the incoming call and each response mode, and if the matching result is matched, responding according to actions in the matched response mode; if the user intention prompt is not matched, playing the converted prompt which is not understood and identified in advance.
Further, in the matching process of the large language model, the content of the incoming call is matched with the intention corresponding to the response mode by combining the intention description corresponding to the response mode.
Further, actions are divided into two types, one type contains a dialing address, and the other type does not contain a dialing address;
When responding to actions containing dialing addresses according to the actions in the matched response mode, dialing the corresponding dialing addresses;
Aiming at actions without dial addresses, configuration items of parameters, parameter descriptions and parameter query text descriptions are also arranged in corresponding response modes, and when the matched result is the response modes, the method also comprises the steps of matching parameters in the call contents through a large language model and combining the parameter descriptions; if the parameter is matched, searching a corresponding dialing address based on the matched parameter to dial; if the parameters are not matched, playing the voice corresponding to the parameter inquiry text description.
Further, the actions containing the dial address include a queue XXXX and a ring group XXXXXX, wherein XXXX represents a specific number; the action without the dial address includes an extension.
Further, the method is implemented based on an ip pbx, and further comprises the following configuration before the method is executed:
(1) Configuring the number of IVR for the identification of external incoming route;
(2) Configuring a sentence breaking timeout period for judging whether a section of speech of a user is ended or not;
(3) Configuring voice broadcasting roles, wherein different roles correspond to different speech speeds, intonation and accents;
(4) Configuring prompt text description of immediate response, which is used for responding to the user immediately after a section of speech of the user is finished;
(5) Configuring a word description of welcome prompts for prompting contents heard by a user after the user calls in the IVR;
(6) The configuration fails to understand the textual description when the user's intent is identified.
An intelligent IVR implementing terminal device comprises a processor, a memory and a computer program stored in the memory and capable of running on the processor, wherein the steps of the method according to the embodiment of the invention are implemented when the processor executes the computer program.
A computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of the method described above for embodiments of the present invention.
The invention adopts the technical scheme and has the advantages of simple configuration, good user experience and high efficiency.
Drawings
Fig. 1 is a flowchart showing an implementation of the first embodiment of the present invention.
Detailed Description
For further illustration of the various embodiments, the invention is provided with the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments and together with the description, serve to explain the principles of the embodiments. With reference to these matters, one of ordinary skill in the art will understand other possible embodiments and advantages of the present invention.
The invention will now be further described with reference to the drawings and detailed description.
Embodiment one:
the embodiment of the invention provides an intelligent IVR implementation method, as shown in figure 1, which comprises the following steps: classifying intentions of incoming calls, and configuring a response mode corresponding to each intention; the configuration items in each response mode include at least intent, intent description, and action. When receiving an incoming call, sending the content of the incoming call and each configured response mode to a large language model together to obtain a matching result of the incoming call and each response mode, and if the matching result is matched, responding according to actions in the matched response mode; if the user intention prompt is not matched, playing the converted prompt which is not understood and identified in advance.
The large language model can be a quantized large language model built in the IPPBX, or any large language model (such as chagpt) passing through the cloud.
In this embodiment, the activities are classified into two types, one type containing dial addresses, such as a queue XXXX, a ring group XXXX, etc., and one type not containing dial addresses, such as an extension. Table 1 describes the response patterns employed in this example.
TABLE 1
The parameters mentioned above may be, in addition to the name, parameters such as telephone, identification card number, company name, etc. that can be associated to a specific dialing address.
Based on the response mode configuration described above, the transfer of the before-market service to the queue 6800, the transfer of the technical support to the queue 6801, the transfer of the after-market service to the response group 6200, and the transfer of the call manager to the extension of the customer manager can be realized.
The two types of actions have different response modes, and are described below.
For actions containing dial addresses, in the matching process of the large language model, only the intention is matched, namely, the intention corresponding to the response mode is combined with the intention description corresponding to the response mode, and the incoming call content is matched with the intention corresponding to the response mode. When responding according to the actions in the matched response mode, the corresponding dialing address is dialed.
For actions without dial addresses, in the matching process of the large language model, matching parameters are needed besides matching intention, namely when the matched result is in such a response mode, the matching process also comprises matching parameters in the call content through the large language model by combining parameter description. If the parameters are matched, the corresponding dialing address is searched for dialing based on the matched parameters, and if the names are matched, the extension numbers corresponding to the names are searched for based on a preset corresponding table of the names and the extension numbers for dialing; if the parameters are not matched, playing the parameters to inquire the voice corresponding to the text description, such as playing to ask you which customer service manager to find.
The method of the embodiment is realized based on the IPPBX, and the method further comprises the following configuration before executing:
(1) Configuring the number of IVR for the identification of external incoming route;
(2) Configuring a sentence breaking timeout period for judging whether a section of speech of a user is ended or not;
(3) Configuring voice broadcasting roles, wherein different roles correspond to different speech speeds, intonation and accents;
(4) And configuring prompt text description of immediate response, which is used for responding to the user immediately after a section of speech of the user is finished, and preventing the user from experiencing unresponsiveness when the IPPBX processes too much in some cases.
(5) Configuring a word description of welcome prompts for prompting contents heard by a user after the user calls in the IVR;
(6) The configuration fails to understand the textual description when the user's intent is identified.
After the configuration is finished, the IPPBX system can do the following preparation work:
(1) According to the voice broadcasting role, the prompt text description which is responded immediately, the text description which welts the prompt, the text description which fails to understand and identify the user intention and the parameter inquiry text description are converted into voice for playing interactive use when the external user is in. The voice-to-text function may be built into the ip pbx or may be converted by invoking the cloud STT service, without limitation.
(2) And the configured multiple groups of response services are converted into prompt languages of the AI model, so that the AI model can understand and match the intention of the return user.
As shown in fig. 1, based on the above method and configuration, the specific implementation flow of this embodiment includes the following steps:
S1: the external number dials the number that the enterprise opens to the outside, such as 400 phones.
S2: ip pbx routes the 400 incoming call to the configured IVR number.
S3: and playing the pre-converted welcome prompt.
S4: waiting for the user to speak, and when the user starts speaking, starting to enter a state of receiving speech.
S5: and the voice is converted into characters in parallel, so that the voice is converted into characters in real time, and the response time of the whole flow is reduced.
S6: and waiting for the sentence breaking timeout period after the user does not speak, and if the sentence breaking timeout period exceeds the timeout period, considering that the user speaks to be finished. Otherwise, continuing to receive the voice.
S7: and sending the incoming call content of the current user and each response mode for realizing the configured configuration to the large language model so as to match the response modes.
S8: judging whether the response mode is matched (namely, the identified intention is configured intention), if so, entering S9; otherwise, playing the converted prompt which fails to understand and identify the intention of the user.
S9: judging whether the matched response mode contains parameters or not, if so, entering S10; otherwise, dialing is carried out according to the dialing address corresponding to the action in the matched response mode.
S10: judging whether the parameters are matched (namely, the identified parameters are configured parameters), if so, searching the corresponding dialing address based on the matched parameters to dial; otherwise, the playing parameter inquires the voice corresponding to the text description.
The embodiment of the invention has the following beneficial effects:
(1) The configuration is simple: the prompt tone does not need to be uploaded, and the key event does not need to be set. One company only needs to configure one IVR number, and multiple IVRs are not required to be created.
(2) The user experience is good: without hearing a large section of key prompting voice, the service to which the incoming user needs to respond is automatically identified in a dialogue mode. Ordinary intelligent IVR also can make prompt query of user intention, based on large model intelligent IVR, without excessive guiding prompt tone, can be a simple sentence "here is XXX please ask me what can help you? ".
(3) The efficiency is high: the user can quickly analyze and acquire the purpose of the user through one-time voice input. For example, the user can say that "I want to find XXX customer service manager" at first, and can immediately transfer the call to customer service manager. Instead of listening to the alert tone first, the user speaks "I want to find a customer service manager" according to the alert tone, the system asks "please ask you what customer service manager to find, please complain about I's name", and the user answers "XXX".
Embodiment two:
The invention also provides an intelligent IVR realizing terminal device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the steps in the method embodiment of the first embodiment of the invention are realized when the processor executes the computer program.
Further, as an executable scheme, the intelligent IVR implementation terminal device may be a computing device such as a desktop computer, a notebook computer, a palm computer, and a cloud server. The intelligent IVR implementation terminal device may include, but is not limited to, a processor, a memory. It will be appreciated by those skilled in the art that the foregoing constituent structure of the intelligent IVR implementation terminal device is merely an example of the intelligent IVR implementation terminal device, and does not constitute limitation of the intelligent IVR implementation terminal device, and may include more or fewer components than the foregoing, or combine some components, or different components, for example, the intelligent IVR implementation terminal device may further include an input/output device, a network access device, a bus, and the embodiment of the present invention does not limit this.
Further, as an executable scheme, the Processor may be a central processing unit (Central Processing Unit, CPU), other general purpose Processor, digital signal Processor (DIGITAL SIGNAL Processor, DSP), application SPECIFIC INTEGRATED Circuit (ASIC), field-Programmable gate array (Field-Programmable GATE ARRAY, FPGA), or other Programmable logic device. The general processor may be a microprocessor or the processor may also be any conventional processor, etc., where the processor is a control center of the intelligent IVR implementation terminal device, and various interfaces and lines are used to connect various parts of the entire intelligent IVR implementation terminal device.
The memory may be used to store the computer program and/or module, and the processor may implement various functions of the terminal device by running or executing the computer program and/or module stored in the memory, and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for a function; the storage data area may store data created according to the use of the cellular phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart memory card (SMART MEDIA CARD, SMC), secure Digital (SD) card, flash memory card (FLASH CARD), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
The present invention also provides a computer readable storage medium storing a computer program which when executed by a processor implements the steps of the above-described method of an embodiment of the present invention.
The modules/units of the intelligent IVR implementation terminal device integration may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as a stand alone product. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a software distribution medium, and so forth.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (7)

1. An intelligent IVR implementation method is characterized by comprising the following steps:
Classifying intentions of incoming calls, and configuring a response mode corresponding to each intention; the configuration items in each response mode include at least intent, intent description, and action;
When receiving an incoming call, sending the content of the incoming call and each configured response mode to a large language model together to obtain a matching result of the incoming call and each response mode, and if the matching result is matched, responding according to actions in the matched response mode; if the user intention prompt is not matched, playing the converted prompt which is not understood and identified in advance.
2. The intelligent IVR implementation method according to claim 1, wherein: in the matching process of the large language model, the content of the incoming call is matched with the intention corresponding to the response mode by combining with the intention description corresponding to the response mode.
3. The intelligent IVR implementation method according to claim 1, wherein: actions are divided into two types, one type contains a dialing address, and the other type does not contain a dialing address;
When responding to actions containing dialing addresses according to the actions in the matched response mode, dialing the corresponding dialing addresses;
Aiming at actions without dial addresses, configuration items of parameters, parameter descriptions and parameter query text descriptions are also arranged in corresponding response modes, and when the matched result is the response modes, the method also comprises the steps of matching parameters in the call contents through a large language model and combining the parameter descriptions; if the parameter is matched, searching a corresponding dialing address based on the matched parameter to dial; if the parameters are not matched, playing the voice corresponding to the parameter inquiry text description.
4. The intelligent IVR implementation method according to claim 3, wherein: the actions containing the dial address include a queue XXXX and a ringing group XXXX, where XXXX represents a specific number; the action without the dial address includes an extension.
5. The intelligent IVR implementation method according to claim 1, wherein: the method is based on an IPPBX implementation and comprises the following configurations before the method is executed:
(1) Configuring the number of IVR for the identification of external incoming route;
(2) Configuring a sentence breaking timeout period for judging whether a section of speech of a user is ended or not;
(3) Configuring voice broadcasting roles, wherein different roles correspond to different speech speeds, intonation and accents;
(4) Configuring prompt text description of immediate response, which is used for responding to the user immediately after a section of speech of the user is finished;
(5) Configuring a word description of welcome prompts for prompting contents heard by a user after the user calls in the IVR;
(6) The configuration fails to understand the textual description when the user's intent is identified.
6. An intelligent IVR realizing terminal device is characterized in that: comprising a processor, a memory and a computer program stored in the memory and running on the processor, which processor, when executing the computer program, carries out the steps of the method according to any one of claims 1 to 5.
7. A computer-readable storage medium storing a computer program, characterized in that: the computer program implementing the steps of the method according to any one of claims 1 to 5 when executed by a processor.
CN202410195587.XA 2024-02-22 2024-02-22 Intelligent IVR implementation method, terminal equipment and storage medium Pending CN117998014A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410195587.XA CN117998014A (en) 2024-02-22 2024-02-22 Intelligent IVR implementation method, terminal equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410195587.XA CN117998014A (en) 2024-02-22 2024-02-22 Intelligent IVR implementation method, terminal equipment and storage medium

Publications (1)

Publication Number Publication Date
CN117998014A true CN117998014A (en) 2024-05-07

Family

ID=90897287

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410195587.XA Pending CN117998014A (en) 2024-02-22 2024-02-22 Intelligent IVR implementation method, terminal equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117998014A (en)

Similar Documents

Publication Publication Date Title
CN110891124B (en) System for artificial intelligence pick-up call
US7460652B2 (en) VoiceXML and rule engine based switchboard for interactive voice response (IVR) services
US7167547B2 (en) Personal calendaring, schedules, and notification using directory data
US6870915B2 (en) Personal address updates using directory assistance data
US5822727A (en) Method for automatic speech recognition in telephony
US7940898B2 (en) Method and system for routing calls based on a language preference
US8625749B2 (en) Content sensitive do-not-disturb (DND) option for a communication system
US6505161B1 (en) Speech recognition that adjusts automatically to input devices
US20080071544A1 (en) Integrating Voice-Enabled Local Search and Contact Lists
JPH0918574A (en) Telephone response device
US20110238414A1 (en) Telephony service interaction management
CA2947242C (en) Voice call diversion to alternate communication method
US7319742B2 (en) Voice information storage and retrieval system and method
US20040161079A1 (en) Methods and systems for routing a call
US20060246891A1 (en) Voice mail with phone number recognition system
US20040193603A1 (en) Technique for effectively searching for information in response to requests in information assistance service
WO2002039422A2 (en) Voice telephone book service system using voice recognition and method thereof
JP2016225740A (en) Speech communication distribution system, call control device and program
CN117998014A (en) Intelligent IVR implementation method, terminal equipment and storage medium
CN115604241A (en) Privacy number playing method, system, computer equipment and storage medium
US7328005B2 (en) Information storage system
CN112887495A (en) Intelligent voice customer service menu sequence dynamic adjusting method
KR102686276B1 (en) System for providing ai based automatic telephone response and method recording the same
CN114598773B (en) Intelligent response system and method
CN108111703A (en) Contact method querying method and telephone exchange server

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