TWM608966U - Temporary credit card credit line adjustment system for artificial intelligence voice - Google Patents

Abstract

A credit card temporary quota adjustment system includes: a processing device that recognizes a text message to be recognized according to a plurality of classification models to generate a user intention, and matches the user intention with a plurality of text response models to generate a text query message; The bank-side server device determines whether to adjust the temporary credit card quota based on a quota adjustment data and a quota adjustment request to generate an adjustment result; and an information transceiver device first converts a voice input message into the text message to be recognized and outputs it to The processing device then sends out a voice inquiry message according to the user's intention and the text inquiry message to obtain the user's quota adjustment request and output it to the bank server device, and generates and sends out a voice reply message according to the adjustment result To users.

Description

Temporary credit card quota adjustment system for artificial intelligence voice

This model relates to an adjustment system, especially a temporary credit card adjustment system for artificial intelligence voice.

Every existing credit card has a corresponding spending limit. When a credit card holder (customer) spends with his credit card and the consumption amount exceeds the limit of his credit card, consumers can dial the customer service line and use interactive voice Interactive Voice Response (IVR) is used to perform layer-by-layer voice menu operations through the buttons of the mobile phone to adjust the temporary limit of the credit card. However, customers often have no patience to operate the layered voice menus, and choose to directly call the customer service staff for assistance, resulting in a large number of customers who choose to ask the customer service staff for assistance.

As a result, on the one hand, it will cause an increase in the workload of the customer service staff of the credit card bank. On the other hand, if customers choose to directly find the customer service staff for assistance, each customer will have to wait a long time on the phone before being The customer service staff answered the call, causing customers to spend more time waiting, and it was not convenient for customers to adjust the temporary credit card limit. According to this, the existing temporary credit card adjustment mechanism is indeed necessary to be improved.

Therefore, the purpose of the present invention is to provide a temporary credit card quota adjustment system for artificial intelligence voice that can immediately adjust the customer's temporary credit card quota without the assistance of customer service personnel.

Therefore, the new credit card temporary quota adjustment system for artificial intelligence voice includes a processing device, a bank-side server device, and an information transceiver device.

The processing device stores a plurality of pre-trained classification models and a plurality of text response models, and receives a text information to be recognized, and recognizes the text information to be recognized according to the classification models to generate a user intention, and Matching the user intention with the text response models to generate a text query message.

The bank-side server device receives a quota adjustment request with a credit card identification code and a temporary quota adjustment value, and stores a quota adjustment data that records multiple customer credit card identification codes, and each customer credit card identification code corresponds to one Credit card adjustable limit, the bank-side server device judges whether to adjust the temporary credit card limit based on the limit adjustment data and the limit adjustment request, and generates and outputs an adjustment result accordingly. The adjustment result contains information about whether the temporary limit can be adjusted in real time. .

The information transceiving device is used to receive a user's voice input information related to the adjustment of the temporary credit card limit, and is communicatively connected to the processing device to receive the user's intention and the text inquiry information, and to the bank-side server device to receive the As a result of the adjustment, the information transceiving device first converts the voice input information into the text information to be recognized and outputs it to the processing device, and then sends out a voice query message according to the user’s intention and the text query message to obtain the user’s corresponding information. The quota adjustment request is output to the bank-side server device, and a voice response message is generated and sent to the user according to the adjustment result.

The effect of the new model is that the temporary credit card quota adjustment system can immediately reply whether the user can adjust the temporary quota of his credit card, and there is no need to wait for the customer service staff to answer the temporary quota adjustment as in the prior art, so the temporary quota of the credit card The adjustment system is more convenient for customers to use.

Referring to FIG. 1, an embodiment of a credit card temporary quota adjustment system for artificial intelligence voice of the present invention includes a processing device 1, a bank-side server device 2, and an information transceiver device 3.

The processing device 1 stores a plurality of pre-trained classification models and a plurality of text response models, and receives a text information Ti to be recognized. The processing device 1 recognizes the text information Ti to be recognized according to the classification models to generate a user intention Ui corresponding to the temporary credit card limit adjustment, and matches the user intention Ui with the text response models to generate a text Ask for information Ta. In this embodiment, the processing device 1 is a natural language processing (NLP) device and includes an intent recognition module 11 and a result matching module 12. The text information Ti to be recognized can be stored in the form of a file, and the format of the file includes but is not limited to text formats such as txt and doc.

The intention recognition module 11 stores the classification models and receives the text information Ti to be recognized, and recognizes the text information Ti to be recognized according to the classification models to generate the user intention Ui. These classification models are classification models for text classification. The classification model may be a neural network-based classification model, or a deep learning-based classification model, and is not limited to this.

The result matching module 12 stores the text response models, and is communicatively connected to the intention recognition module 11 to receive the user intention Ui, and match the user intention Ui with the text response models to generate the text query Information Ta.

The bank-side server device 2 stores a quota adjustment data, and receives a quota adjustment request Qa with a credit card identification code and a temporary quota adjustment value. The limit adjustment data records multiple customer credit card identification codes, and each customer credit card identification code corresponds to an adjustable credit card limit. The quota adjustment request Qa is related to a user's request for pre-adjusting the temporary quota of the credit card. The bank-side server device 2 determines whether to adjust the temporary credit card quota of the user according to the quota adjustment data and the quota adjustment request Qa, and generates and outputs an adjustment result Ar to the information transceiver device 3 accordingly. The adjustment result Ar contains information on whether the temporary quota can be adjusted in real time. When the adjustment result Ar contains information that the temporary quota cannot be adjusted in real time, the information transceiver device 3 sends a notification command Ni containing the adjustment result Ar and the quota adjustment request Qa to the bank server 2. The bank server device 2 also sends an audit result notification message to a mobile device of the user at a predetermined time according to the notification instruction Ni. In this embodiment, the bank server 2 includes a decision module 21 and a notification module 22.

The decision module 21 stores the quota adjustment data, and is communicatively connected to the information transceiver 3 to receive the quota adjustment request Qa, and determines whether to adjust the temporary credit card quota according to the quota adjustment data and the quota adjustment request Qa to generate the Adjust the result Ar.

The notification module 22 is communicatively connected to the information transceiver device 3 to receive the notification instruction Ni, and according to the notification instruction Ni, sends the audit result notification information to the mobile device of the user at the predetermined time.

The information transceiving device 3 is used to receive a voice input information Vi related to the adjustment of the temporary credit card limit of the user, and is communicatively connected to the processing device 1 to receive the user's intention Ui and the text inquiry information Ta, and communicatively connected to the bank terminal The servo device 2 receives the adjustment result Ar. The information transceiving device 3 first converts the voice input information Vi into the text information Ti to be recognized and outputs it to the processing device 1, and then sends a voice query information Va according to the user's intention Ui and the text query information Ta to obtain correspondingly The quota adjustment request Qa of the user is output to the bank-side server device 2. The information transceiving device 3 also outputs the adjustment result Ar to the processing device 1, and the processing device 1 matches the adjustment result Ar with the text response models to generate and output a text result message Tr to the information transceiving device 3. The message transceiving device 3 converts the text result message Tr into a voice response message Vr and sends it to the user.

In this embodiment, the information transceiving device 3 includes a voice recognition module 31, a voice synthesis module 32, and an interactive voice response (Interactive Voice Response, IVR) module 33. The speech recognition module 31 is an Automatic Speech Recognition (ASR) module for performing speech-to-text recognition (Speech To Text, STT). The speech synthesis module 32 is used for text to speech recognition (Text to Speech, TTS).

The voice recognition module 31 receives the voice input information Vi, and converts the voice input information Vi into the to-be-recognized text information Ti and outputs it. The voice synthesis module 32 receives the text query information Ta and the text result information Tr, and converts the text query information Ta and the text result information Tr into the voice query information Va and the voice response information Vr, respectively, and outputs them to the voice query information Va and the voice response information Vr. The interactive voice response module 33 is sent by the interactive voice response module 33 to the user.

The interactive voice response module 33 is communicatively connected to the voice recognition module 31, the voice synthesis module 32, the processing device 1 and the bank-side server device 2 to obtain the voice input information Vi and output it to the voice recognition module The group 31 receives the text information Ti to be recognized from the voice recognition module 31 and outputs it to the processing device 1. The interactive voice response module 33 receives the text query information Ta and the text result information Tr from the processing device 1 and outputs them to the voice synthesis module 32 to obtain the voice query information Va from the voice synthesis module 32 And the voice reply message Vr is sent out. The interactive voice response module 33 also receives the user intention Ui from the processing device 1 and the adjustment result Ar from the bank-side server device 2, and outputs the adjustment result Ar to the processing device 1, and the amount The adjustment request Qa is output to the bank-side server 2. It should be noted that, in this embodiment, the interactive voice response module 33 includes a voice processor 331 and a logic processor 332. The interactive voice response module 33 is a prior art, and the configuration and operation of the voice processor 331 and the logic processor 332 are well known by those skilled in the art. For the sake of brevity, it will not be repeated here. .

2 to 4, and then a detailed description of a credit card temporary quota adjustment method executed by the credit card temporary quota adjustment system. The method for adjusting the temporary quota of a credit card includes steps 40 to 57. When the user pre-adjusts the temporary limit of his credit card, the user dials the customer service line of his credit card company and starts a conversation, so that the information transceiving device 3 receives the voice input information Vi (that is, what the user said) .

In step 40, the voice processor 331 of the interactive voice response module 33 obtains the voice input information Vi from the user (for example, "to increase the temporary credit card limit"), and outputs the voice input information Vi to The voice recognition module 31.

In step 41, the voice recognition module 31 converts the voice input information Vi into the to-be-recognized text information Ti, and outputs the to-be-recognized text information Ti to the logical processor 332 of the interactive voice response module 33 The intention recognition module 11.

In step 42, the intention recognition module 11 recognizes the to-be-recognized text information Ti according to the classification models to generate the user intention Ui (ie, "increasing the temporary credit card limit").

In step 43, the result matching module 12 matches the user intention Ui and the text response models to generate the text query message Ta (for example, "Please tell me the ID number or card number, and what you need to adjust) The quota and purpose of ""), and output the text query information Ta to the speech synthesis module 32 through the logic processor 332.

It should be noted that, in other embodiments, if the interactive voice response module 33 has obtained the user’s identity card number or card number after the user dials the customer service line to enter the line, then the text in step 43 The inquiry information Ta does not need to include "Please state your ID number or card number". In addition, in step 43, the result matching module 12 can first generate only the text query message Ta including "Please tell me the ID number or card number" for the information transceiver module 3 to send out to query the user accordingly , To perform identity verification on the user, and after the identity verification is successful, the result matching module 12 then generates only the text query message Ta including "the amount and purpose you need to adjust" for subsequent operations, that is, step The implementation aspect of 43 is not limited to this embodiment.

In step 44, the voice synthesis module 32 converts the text query information Ta into the voice query information Va, and outputs the voice query information Va to the voice processor 331, so that the voice processor 331 generates the voice query information Va (ie, " Please state the ID number and card number, as well as the amount and purpose you need to adjust") and send it to the user.

In step 45, the voice processor 331 responds to the voice query message Va and correspondingly receives a voice message to be authenticated Vid that contains the user's ID number and/or card number, and a voice message that contains the temporary quota adjustment. Voice adjustment information Vam that is worth and its purpose (for example, "Increase 200,000 to buy a car").

In step 46, the voice recognition module 31 converts the to-be-authenticated voice information Vid and the voice adjustment information Vam into a to-be-authenticated text information Tid and a to-be-recognized text adjustment information Tam, respectively, and combines the to-be-authenticated text information Tid and The text adjustment information Tam to be recognized is output to the logic processor 332.

In step 47, the logic processor 332 obtains a customer data from a host (not shown) of the bank-side server device 2, and performs identity verification on the user based on the customer data and the text information Tid to be authenticated to Determine whether the user identity is correct. If the judgment is no, go to step 48; if the judgment is yes, go to step 49.

In step 48, the interactive voice response module 33 sends a voice message indicating that the identity verification fails to the user.

In step 49, the logic processor 332 outputs the to-be-recognized text adjustment information Tam to the intention recognition module 11.

In step 50, the intention recognition module 11 recognizes the to-be-recognized text adjustment information Tam according to the classification models to generate a user adjustment intention Uia containing the temporary quota adjustment value and its purpose, and output it to the logic Processor 332.

In step 51, the logic processor 332 obtains the quota adjustment request Qa of the user according to the user's adjustment intention Uia and the credit card identification code (that is, the quota adjustment request Qa has the credit card identification code and the user wants to increase 200,000 to buy a car), and output the quota adjustment request Qa to the decision-making module 21.

It should be noted that the credit card identification code can be provided by the user, or the interactive voice response module 33 itself stores the credit card identification code of each user (which corresponds to the phone number of each user). When the user dials the customer service line, the interactive voice response module 33 can correspondingly obtain the phone number currently used by the user, and obtain the corresponding credit card identification code according to the phone number.

In step 52, the decision module 21 determines whether to adjust the temporary credit card quota according to the quota adjustment data and the quota adjustment request Qa to generate the adjustment result Ar, and outputs the adjustment result Ar to the logic processor 332. If the judgment is yes (that is, the adjustment result Ar contains the information that the temporary limit of the credit card can be adjusted in real time), go to step 53 to step 55; if the judgment is no (that is, the adjustment result Ar contains the temporary limit of the credit card). If the quota cannot be adjusted in real time), proceed to step 53 to step 57.

In step 53, the logic processor 332 outputs the adjustment result Ar to the result matching module 12.

In step 54, the result matching module 12 matches the adjustment result Ar and the text response models to generate the text result information Tr, and output the text result information Tr to the logic of the information transceiver 3 Processor 332.

For example, when the judgment result of step 51 is yes, the text result information Tr is "the temporary credit card limit has been increased by 200,000", and when the judgment result of step 51 is no, the text result information Tr is "temporary credit card The quota must be reviewed for one working day, and you will be notified of the review result in a short message."

In step 55, the information transceiving device 3 converts the text result information Tr into the voice response information Vr and sends it to the user.

In detail, the logic processor 332 first transmits the text result information Tr to the speech synthesis module 32, and then the speech synthesis module 32 converts the text result information Tr into the voice response information Vr and outputs it to the speech The processor 331, and finally, the voice processor 331 sends the voice reply message Vr to the user.

In step 56, when the determination result of step 51 is no, the logic processor 332 sends the notification instruction Ni carrying the adjustment result Ar and the quota adjustment request Qa to the notification module 22 of the bank-side server device .

In step 57, the notification module 22 sends the audit result notification information to the mobile device of the user at the predetermined time (for example, two days later) according to the notification instruction Ni.

For example, the notification module 22 displays the notification instruction Ni to the customer service staff of the credit card company to notify the customer service staff to perform a detailed manual review, and use the notification module 22 to send the review result notification message after the review is completed. To the mobile device of the user.

In summary, with the credit card temporary quota adjustment system executing the credit card temporary quota adjustment method, the user can immediately reply whether the user can adjust the temporary quota of his credit card, and there is no need to wait for the customer service staff to answer the temporary quota as in the prior art. As a result, the temporary credit card quota adjustment system of the present invention is more convenient for customers to use and can reduce the workload of customer service staff. In addition, with the temporary credit card quota adjustment system, the user does not need to operate through a layer of voice menus as in the prior art, thereby saving the user's time and quickly completing the quota adjustment service.

However, the above are only examples of the present model, and should not be used to limit the scope of implementation of the present model, all simple equivalent changes and modifications made in accordance with the patent scope of the present model application and the contents of the patent specification still belong to This new patent covers the scope.

1: Processing device 11: Intent recognition module 12: Result matching module 2: Bank-side server device 21: Decision Module 22: Notification module 3: Information transceiver 31: Voice recognition module 32: Speech synthesis module 33: Interactive voice response module 331: Voice Processor 332: logical processor 40~57: Step Ar: Adjust the result Ni: Notification instruction Qa: Quota adjustment request Ta: text asking for information Tam: To be recognized text adjustment information Ti: Text information to be recognized Tid: Text message to be authenticated Tr: text result information Ui: user intent Uia: User adjustment intention Va: Voice asking for information Vam: Voice adjustment information Vi: Voice input information Vid: Voice message to be authenticated Vr: voice reply message

The other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, among which: Fig. 1 is a block diagram illustrating an embodiment of the credit card temporary limit adjustment system for artificial intelligence voice according to the present invention; and Figures 2 to 4 are flowcharts illustrating a method for adjusting a temporary credit card limit performed by this embodiment.

1: Processing device

11: Intent recognition module

12: Result matching module

2: Bank-side server device

21: Decision Module

22: Notification module

3: Information transceiver

31: Voice recognition module

32: Speech synthesis module

33: Interactive voice response module

331: Voice Processor

332: logical processor

Ar: Adjust the result

Ni: Notification instruction

Qa: Quota adjustment request

Ta: text asking for information

Tam: To be recognized text adjustment information

Ti: Text information to be recognized

Tid: Text message to be authenticated

Tr: text result information

Ui: user intent

Uia: User adjustment intention

Va: Voice asking for information

Vam: Voice adjustment information

Vi: Voice input information

Vid: Voice message to be authenticated

Vr: voice reply message

Claims (8)

A temporary credit card quota adjustment system for artificial intelligence voice, including: A processing device storing a plurality of pre-trained classification models and a plurality of text response models, and receiving a text information to be recognized, and recognizing the text information to be recognized according to the classification models to generate a user intention, And matching the user's intention with the text response models to generate a text query message; A bank-side server device receives a quota adjustment request with a credit card identification code and a temporary quota adjustment value, and stores a quota adjustment data that records multiple customer credit card identification codes, each corresponding to the customer credit card identification code A credit card adjustable limit. The bank-side server device determines whether to adjust the temporary credit card limit based on the limit adjustment data and the limit adjustment request, and generates and outputs an adjustment result accordingly. The adjustment result contains whether the temporary limit can be adjusted in real time Information; and An information transceiver device for receiving a user's voice input information related to the adjustment of the temporary credit card limit, and communicating with the processing device to receive the user's intention and the text inquiry information, and communicating with the bank server device to receive As a result of the adjustment, the information transceiving device first converts the voice input information into the text information to be recognized and outputs it to the processing device, and then sends out a voice query message according to the user's intention and the text query message, so as to correspondingly obtain the user’s information The quota adjustment request is output to the bank-side server device, and a voice response message is generated and sent to the user according to the adjustment result. The credit card temporary quota adjustment system described in claim 1, wherein: When the information transceiving device sends out the voice inquiry message, and correspondingly receives a voice adjustment message containing the temporary quota adjustment value and its purpose from the user, the information transceiving device converts the voice adjustment information into a to-be-recognized The text adjustment information is output to the processing device, The processing device recognizes the text adjustment information to be recognized according to the classification models to generate a user adjustment intention containing the temporary quota adjustment value and its purpose, and output it to the information transceiver device, and The information transceiver device obtains and outputs the quota adjustment request of the user according to the user's adjustment intention and the credit card identification code. The credit card temporary quota adjustment system according to claim 1, wherein, when the adjustment result contains information that the temporary quota cannot be adjusted in real time, the information transceiver device also sends a message containing the adjustment result and the quota adjustment request A notification instruction is sent to the bank-side server device, and the bank-side server device also sends an audit result notification message to a mobile device of the user at a predetermined time according to the notification instruction. The credit card temporary quota adjustment system according to claim 3, wherein the bank-side server device includes: A decision-making module stores the quota adjustment data, and is communicatively connected to the information transceiver to receive the quota adjustment request, and determines whether to adjust the temporary credit card quota according to the quota adjustment data and the quota adjustment request to generate the adjustment result, and A notification module is communicatively connected to the information transceiving device to receive the notification instruction, and sends the audit result notification information to the mobile device of the user at the predetermined time according to the notification instruction. The credit card temporary quota adjustment system according to claim 1, wherein the processing device includes: An intention recognition module that stores the classification models and receives the text information to be recognized, and recognizes the text information to be recognized according to the classification models to generate the user intention, and A result matching module stores the text response models, and is communicatively connected to the intention recognition module to receive the user intention, and matches the user intention with the text response models to generate the text query message. The credit card temporary quota adjustment system according to claim 1, wherein the information transceiving device also outputs the adjustment result to the processing device, and the processing device also matches the adjustment result with the text response models to generate a combination Outputting a text result message to the information transceiver device, and the message transceiver device also converts the text result message into the voice response message. The credit card temporary quota adjustment system according to claim 6, wherein the information transceiving device includes: A voice recognition module that receives the voice input information, converts the voice input information into the text information to be recognized and outputs it, A voice synthesis module that receives the text query information and the text result information, and converts the text query information and the text result information into the voice query information and the voice response information and outputs them, and An interactive voice response module, communicatively connected to the voice recognition module, the voice synthesis module, the processing device and the bank-side server device, used to obtain the voice input information and output to the voice recognition module, and receive from The text information to be recognized of the speech recognition module is output to the processing device, and the text query information and the text result information from the processing device are received and output to the speech synthesis module to obtain information from the speech synthesis module. The voice inquiry message and the voice reply message of the group are sent out. The interactive voice response module also receives the user’s intention from the processing device and the adjustment result from the bank-side server device, and outputs the adjustment result to the Processing device, and output the quota adjustment request to the bank-side server device. The credit card temporary quota adjustment system according to claim 1, wherein the processing device is a natural language processing device.

Images