TWM578858U - Cross-channel artificial intelligence dialogue platform - Google Patents

Abstract

The cross-channel AI dialogue platform includes three internal servers. The voice input interface of the first internal server receives the first voice signal. The personal hiding module of the second internal server deletes the personal data in the first voice signal to generate a second voice signal. The voice-to-text interface and the semantic interface of the first internal server respectively convert the second voice signal into a first text signal and obtain an intent signal according to the first text signal. The dialog module and the application module of the third internal server respectively generate a reply signal and a control command according to the intention signal. The text-to-speech interface of the first internal server converts the reply signal into a third voice signal. The output interface outputs a third voice signal and a control command.

Description

Cross-channel artificial intelligence dialogue platform

This creation is about a dialogue platform and a dialogue platform, especially a cross-channel artificial intelligence dialogue platform and its operation method.

With the popularity of various digital marketing channels, people encounter any problems in the transaction, and often hope to get a reply in the first time.

However, for financial institutions that provide customer service, increasing the number of customer service personnel will inevitably lead to a significant increase in labor costs. In addition, it takes a certain amount of time to train a good customer service staff. When a sudden influx of a large number of customers comes in, the customer service personnel who can cope with various customer problems are often overwhelmed. The customer service staff who arrive late may not be able to satisfy all kinds of customers. problem. Therefore, for financial institutions that are unable to provide good services, the public's evaluation will be significantly reduced, which will affect the public's trust and willingness to participate in other projects of the financial institution.

In view of this, this creation proposes a cross-channel artificial intelligence (AI) conversational platform. The path includes: a digital channel, a customer service center, and a business unit. Through the introduction of speech recognition system, artificial intelligence dialogue system and artificial intelligence dialogue background, combined with a new generation of conversational artificial intelligence technology, including: Natural Language Processing (NLP), dynamic learning mechanism, multi-round situational dialogue design and The dynamic information collection mechanism establishes a customer dialogue analysis background to enhance the user experience and market influence of the digital channel.

A cross-channel artificial intelligence dialogue platform according to an embodiment of the present invention comprises: a first internal server, comprising: a voice input interface, a voice-to-text interface, a semantic recognition interface, a text-to-speech interface, and an output The interface, wherein the voice input interface is configured to receive a first voice signal; a second internal server is connected to the first internal server, and includes: a client audio database for storing a plurality of audio files, The content of the audio file corresponds to a plurality of personal data; and a hidden module is electrically connected to the customer audio database, and the hidden module is configured to divide the first voice signal into a plurality of voice segments, and When the hidden module determines that any one of the voice segments meets any of the audio files, the hidden module deletes the audio information corresponding to the voice segment from the first voice signal. And returning, by the first voice signal, the first voice signal of the audio data of the voice segment to the first internal server; wherein the first internal The voice-to-text interface of the server is configured to generate a first text signal according to the second voice signal; the semantic interface is configured to generate an intent signal according to the first text signal; and a third internal server, the communication connection is An internal server includes: a dialog module for selectively generating a reply signal according to the intent signal; and an application module for generating a control command corresponding to the intent analysis signal; wherein the first The text-to-speech interface of the internal server is configured to generate a third voice signal according to the reply signal; the output interface of the first internal server is configured to output the third voice signal and the control command.

A method for operating a cross-channel artificial intelligence dialogue platform according to an embodiment of the present invention comprises: receiving a first voice signal by a voice input interface of a first internal server; and serving as a client of a second internal server The audio database stores a plurality of audio files, and the contents of the audio files respectively correspond to a plurality of personal data, wherein the second internal server communicates with the first internal server; and the second internal server is hidden by a second internal server The module divides the first voice signal into a plurality of voice segments, wherein the resource hiding module is electrically connected to the client audio database, and when the resource hiding module determines any one of the voice segments When the audio file is matched, the audio module deletes the audio information corresponding to the voice segment from the first voice signal; and the second voice signal is returned to the a voice-to-text interface of the first internal server, wherein the second voice signal is used to delete the first voice signal of the audio data of the voice segment; The second voice signal generates a first text signal; the semantic interface of the first internal server generates an intent signal according to the first text signal; and the dialog module of the third internal server generates the image according to the intent signal a reply signal, wherein the third internal server is communicatively coupled to the first internal server; and a text-to-speech interface of the first internal server generates a third voice signal according to the reply signal; and one of the three servers The application module generates a control command according to the intent signal; and outputs the third voice signal and the control command by using an output interface of the first internal server.

The above description of the disclosure and the following description of the embodiments are intended to illustrate and explain the spirit and principles of the present invention, and to provide further explanation of the scope of the patent application of the present invention.

The detailed features and advantages of the present invention are described in detail below in the embodiments, which are sufficient to enable any skilled artisan to understand the technical contents of the present invention and implement it according to the contents, the scope of the patent application and the drawings. Anyone familiar with the relevant art can easily understand the purpose and advantages of this creation. The following examples are intended to further illustrate the scope of this creation, but do not limit the scope of the creation in any way.

Please refer to FIG. 1 , which illustrates an architectural diagram 100 of a cross-channel artificial intelligence dialog platform of an embodiment of the present invention. The cross-channel artificial intelligence dialogue platform 100 includes a first internal server 2, a second internal server 4, and a third internal server 6. As shown in FIG. 1, the second internal server 4 and the third internal server 6 are respectively communicably connected to the first internal server 2. In addition, the components in the first internal server 2 are each in communication with the client device 91, the first external server 93, the second external server 95, and the third external server 97.

In practice, the first internal server 2, the second internal server 4, and the third internal server 6 are, for example, Blade Servers, Rack Servers, or uprights configured in a financial institution room. The server (Pedestal Servers), this creation does not limit the hardware types of the first, second and third internal servers 2, 4 and 6.

The first internal server 2, the second internal server 4, and the third internal server 6 each have a memory to implement various functions described later. The above memory may be, for example, a random access memory, a read only memory, or a flash memory. In an embodiment, the first internal server 2, the second internal server 4, and the third internal server 6 further include communication devices supporting wired networks, wireless networks, mobile networks, and/or wireless communications. In one embodiment, the first internal server 2, the second internal server 4, and the third internal server 6 each include a processing circuit that performs the functions described below. The processing circuit is, for example, a microcontroller, a microprocessor, a processor, a central processing unit (CPU), a digital signal processor, a special application integrated circuit. (application specific integrated circuit, ASIC), digital logic circuit, field programmable gate array (FPGA) and / or other hardware components with arithmetic processing functions, this creation does not deal with the hardware type of the circuit Limited.

Please continue to refer to Figure 1. The first internal server 2 includes a voice input interface 21, a voice-to-text interface 23, a semantic recognition interface 25, a text-to-speech interface 27, and an output interface 29. The voice input interface 21 is communicatively coupled to the client device 91. The client device 91 is, for example, a smart phone or a tablet computer with a mobile banking app (application) installed by the user, or a smart speaker or a smart robot of a smart branch counter, etc., for the user, The conversation is made with the client device 91. In practice, the sound emitted by the user will be generated by the receiver (eg, microphone) of the client device 91 to generate the first voice signal, and then the first voice signal is sent by the communication component of the client device 91 to the first internal server. 2 voice input interface 21. In short, when the user needs to perform the financial related operation, the client device 91 can directly speak to generate the first voice signal, and then the first voice signal will be sent to the voice input interface 21 for processing.

Please refer to the second internal server 4 of FIG. 1 , which includes a customer audio database 41 and a personal hiding module 43 electrically connected to each other. The customer audio database 41 stores a plurality of audio files, and the contents of the audio files correspond to a plurality of personal data. In practice, the second internal server may further include a dynamic information learning module, for example, the manual service recording record of the financial institution is used as the training data in advance, and is recorded by the machine learning method from the customer service recording record. The audio segments belonging to the customer's personal assets are automatically identified, and then the audio segments are stored in the customer audio library 41. The dynamic information learning module can further update the records in the customer audio database according to the first voice signal obtained by the voice input interface 21, which is not limited in this creation.

Please continue to refer to the second internal server 4 of FIG. The personal hiding module 43 is electrically connected to the customer audio database 41 and is communicatively coupled to the voice-to-text interface 23 of the first internal server 2. The personal hiding module 43 is configured to divide the first voice signal into a plurality of voice segments, and when the personal hiding module 43 determines that any of the voice segments meets the plurality of audio files stored in the customer information database 41 In either case, the privilege hiding module 43 deletes the audio information corresponding to the voice segment from the first voice signal, and transmits the first voice signal of the audio data of the deleted voice segment as the second voice signal to the second voice signal. The voice of the first internal server 2 is changed to the text interface 23. When the collocation module 43 performs the comparison, for example, a fuzzy comparison algorithm can be employed. In addition, when the compared user's personal resources are divided into a plurality of voice segments, the collateral hiding module 43 reorganizes the voice segments with the user's personal resources to extract the complete audio belonging to the user's personal resources. data. Through the processing mechanism of the above-mentioned privilege hiding module 43, the privacy data belonging to the user's individual can be restricted to the first internal server 2 and the second internal server 4 provided in the computer room of the financial institution, and the subsequent voice is When identifying, the user's personal resources will not be leaked to the Internet.

Please refer to Figure 1. The voice-to-text interface 23 of the first internal server 2 is respectively connected to the personal hiding module 43 of the second internal server 4 and the first external server 93, and the semantic interface 25 is communicatively coupled to the second external server 95. The voice-to-text interface 23 generates a first text signal according to the second voice signal, and the semantic recognition interface 25 generates an intent signal according to the first text signal. In other words, the voice-to-text interface 23 converts the voice data including the user's personal assets into characters, and the semantic recognition interface 25 interprets the user's intention from the text. For example, when the first text signal is: "I want to transfer one dollar", the semantic recognition module 25 can learn from the two intentions that "the user wants to transfer money" and the "transfer amount is one dollar" . In practice, the voice-to-text interface 23 and the semantic recognition interface 25 are, for example, an Application Programming Interface (API), and the first external server 93 is, for example, a Google Cloud voice-to-text (STT) external servo. Device. The second external server 95 is, for example, an IBM Watson external server, and provides various Watson cognitive computing services, including a natural language processing (NLP) service for determining customer intent, through a sentence pattern (Pattern). ) Machine learning mechanism to improve the accuracy of semantic understanding.

Please refer to Figure 1. The third internal server 6 includes a dialog module 61 and an application module 63 that are communicatively coupled to the semantic recognition interface 25. The dialog module 61 selectively generates a reply signal based on the intent signal. The application module 63 can generate control instructions corresponding to the intent analysis signals. In practice, the dialog module 61 of the third internal server 6 can provide a dynamic learning mechanism through the machine learning model, thereby greatly improving maintenance efficiency. The dialog module 61 has a multi-round situational dialogue design. In practice, for example, by training the manual customer service record to be analyzed in advance, the dialog analysis model is trained on the Watson platform, and the trained dialog analysis model is stored in the database of the dialog module 61, thereby enabling the dialog module 61 to provide Interactive contextual dialogue design with Context coherence. For example, when the user says "I want to transfer a dollar", the dialog module 61 obtains the "user wants to transfer" and the "transfer amount is one dollar" from the semantic recognition interface 25. In addition to the intentional intent signal, the user can also provide the user with a reply signal such as "inquiry to the account of the user to transfer the account" and "inquiry to the account that the user wants to transfer the account" to facilitate the client device. The Mobile Bank App running on 91 can collect enough information to complete subsequent transfer operations. In addition, the dialog module 61 has a dynamic information collection mechanism, which can quickly set parameters for rapid deployment and improve the user experience. When the dialog module 61 cannot recognize the user's intention signal, the dialog module 61 can be transferred to the manual customer service system, and the online customer service personnel can respond to the user's question.

Please continue to refer to Figure 1. The text-to-speech interface 27 of the first internal server 2 is communicatively coupled to the dialog module 61 of the third internal server 6 and the third external server 97, respectively. The text-to-speech interface 27 is configured to generate a third voice signal according to the reply signal; in other words, the reply generated by the dialog module 61 is converted into a computer voice that the user can understand, and then output by the output interface 29 of the first internal server 2 The third voice signal is sent to the client device 91 so that the speaker or speaker of the client device 91 plays the third voice signal for the user to listen to. In practice, the third external server 97 is, for example, an external server of the Institute of Text-to-Speech Web Service, which can provide a Text To Speech (TTS) web service, and the server provides SOAP (Simple). Object Access Protocol) A network service that converts input text into speech for output. It should be noted that, although the first, second, and third external servers 93, 95, and 97 of the present invention are in the above embodiment, the first internal server 2 is connected to each cloud service through the Internet. However, in another embodiment, the external server may also purchase a server having a text-to-speech function and a semantic understanding function and set up the server at the local end, and the creation does not limit the first to third external portions. Servers 93~97 must be connected to the cloud to achieve the above functions.

Please continue to refer to Figure 1. The output interface 29 of the first internal server 2 is communicatively coupled to the application module 63 of the third internal server 6 and the client device 91. The output interface 29 outputs a control command generated by the application module 63 in addition to outputting the third voice signal. The control command is, for example, an instruction to control the mobile banking App to complete the transfer operation.

According to the content of the cross-channel artificial intelligence dialogue platform 100 described above, the cross-channel artificial intelligence dialogue platform 100 can be communicatively connected to the smart phone of the user or the smart speaker of the smart branch counter as needed. Thereby, the user can complete the financial transaction operation item to be performed through the dialogue with the client device 91.

Please refer to Figure 1 and Figure 2 together. FIG. 2 is a diagram showing the operation method of the cross-channel artificial intelligence dialogue platform according to an embodiment of the present invention, which is applicable to the foregoing cross-channel artificial intelligence dialogue platform 100. Referring to step S11, the voice input interface 21 receives the first voice signal. In detail, the client device 91 transmits the voice of the user by wired or wireless communication, and is received by the voice input interface 21 of the first internal server 2. Referring to step S21, the personal hiding module 43 of the second internal server 4 divides the first voice signal into a plurality of voice segments. Referring to step S23, the personal hiding module 43 compares the audio segment with the audio file in the client audio library 41. Referring to step S25, the personal hiding module 43 determines whether any of the voice segments meet any of the audio files. If yes, proceed to step S27. Otherwise, it returns to step S23. Referring to step S27, the personal hiding module 43 returns a second voice signal S27, and the second voice signal is used to delete the audio data of the voice segment (for example, a waveform signal representing the audio of the user's personal resources). A voice signal. Referring to step S13, the voice-to-text interface 23 of the first internal server 2 generates a first text signal according to the second voice signal. Specifically, the audio data that does not include the user's personal data is converted into text data. Referring to step S15, the semantic recognition interface 25 of the first internal server 2 generates an intent signal S15 according to the first text signal, for example, the text signal is analyzed by the semantics of the cloud service to obtain the user's intention. Referring to step S31, the dialog module 61 of the third internal server 6 generates a reply signal according to the intent signal. In other words, the dialog module 61 can distinguish the user's intention from the reply signal to give a personalized service or response. . In practice, if the customer's problem is not within the response range of the dialog module 61, the dialog module 61 can be transferred to the manual customer service system for subsequent customer service. Referring to step S17, the text-to-speech interface 27 of the first internal server 2 generates a third voice signal S17 according to the reply signal, which converts the response given by the system to the user into a computer voice audible to the user. Referring to step S33, the application module 63 generates a control command S33 according to the intention signal. The control command is used to perform an operation corresponding to the user's voice on the client device. Referring to step S35, the output interface 29 outputs a third voice signal and a control command to the client device 91, for example, playing back the system to the user for listening, and implementing a dialogue with the user to obtain the operation that the user wants to perform. Other parameters, or execute control commands to complete the financial transaction operation items that the user wants.

In summary, the cross-channel AI dialogue platform disclosed in the present application provides a service for the financial transaction operation by providing a dialogue with the user, so that the customer feels the best experience and service, and can prevent the user's individual. Privacy information is leaked to the cloud to protect users' security. In addition, the cross-channel AI dialogue platform can be connected to the mobile banking app, the smart branch counter or the smart personal wealth management service, which can reduce the labor required for the financial institution to hire and train the personnel who can provide the above financial services. With time cost.

Although the present invention has been disclosed above in the foregoing embodiments, it is not intended to limit the present invention. The changes and refinements that are made without departing from the spirit and scope of this creation are within the scope of patent protection of this creation. Please refer to the attached patent application scope for the scope of protection defined by this creation.

100‧‧‧cross-channel artificial intelligence dialogue platform  2‧‧‧First internal server  21‧‧‧Voice input interface  23‧‧‧Voice to text interface  25‧‧‧Speech Identification Interface  27‧‧‧Text-to-speech interface  29‧‧‧Output interface  4‧‧‧Second internal server  41‧‧‧Customer Audio Library  43‧‧‧ hidden modules  6‧‧‧ Third Internal Server  61‧‧‧Dialog Module  63‧‧‧Application Module  91‧‧‧Client device  93‧‧‧First external server  95‧‧‧Second external server  97‧‧‧ Third external server  S11~S35‧‧‧Steps  

1 is a block diagram of a cross-channel artificial intelligence dialog platform according to an embodiment of the present invention.  2 is a flow chart of a method for operating a cross-channel artificial intelligence dialogue platform according to an embodiment of the present invention.  

Claims (5)

A cross-channel artificial intelligence dialogue platform includes: a first internal server, comprising: a voice input interface, a voice-to-text interface, a semantic recognition interface, a text-to-speech interface, and an output interface, wherein the voice input interface The first internal server is configured to receive the first internal server, and the first internal server is connected to the first internal server, and includes: a customer audio database for storing a plurality of audio files, wherein the contents of the audio files respectively correspond to a plurality of personal data; and a hidden module electrically connected to the customer audio database, the hidden module is configured to divide the first voice signal into a plurality of voice segments, and when the hidden module is used When it is determined that any one of the voice segments meets any of the audio files, the hidden module deletes the audio information corresponding to the voice segment from the first voice signal, and the voice segment is deleted. The first voice signal of the audio data is transmitted back to the first internal server as a second voice signal; wherein the voice of the first internal server is converted to text The interface is configured to generate a first text signal according to the second voice signal; the semantic interface is configured to generate an intent signal according to the first text signal; and a third internal server to communicate with the first internal server, including: a dialog module for selectively generating a reply signal according to the intent signal; and an application module for generating a control command corresponding to the intent analysis signal; wherein the text of the first internal server is The voice interface is configured to generate a third voice signal according to the reply signal; the output interface of the first internal server is configured to output the third voice signal and the control command. The cross-channel artificial intelligence dialogue platform according to claim 1, wherein the voice-to-text interface is connected to a Google Cloud voice-to-text external server, and the semantic recognition interface is connected to the IBM Watson external server, and The text-to-speech interface is connected to the external server of the ITRI text-to-speech web service. The cross-channel artificial intelligence dialogue platform according to claim 1, wherein the dialog module is further connected to an online customer service system, and when the dialogue module cannot distinguish the intention signal, the dialogue module forwards the intention signal To the online customer service system. The cross-channel artificial intelligence dialogue platform according to claim 1, wherein the second internal server further comprises a dynamic information learning module, wherein the dynamic information learning module is configured to record from a customer service and learn from a machine learning manner. The customer service recording records the audio files associated with the personal data and stores the audio files to the customer audio database. The cross-channel artificial intelligence dialogue platform according to claim 1, wherein when the resource hiding module determines that the plurality of voice segments in the first voice signal respectively meet the plurality of audio files in the client audio database, The cryptographic module is further configured to reassemble the voice segments to retrieve a complete profile of the personal data.