KR102597261B1 - A system that can introduce artificial intelligence-based voice recognition service without modifying the existing in-house phone system - Google Patents

Abstract

The present invention relates to an artificial intelligence-based voice recognition providing system for providing an artificial intelligence-based voice recognition function to an Internet phone system. According to the present invention, the voice recognition providing system includes: an in-house phone system including an Internet phone system; a relay telephone system configured to receive a call connection request from a calling terminal to the Internet telephone system and perform a call connection to the Internet telephone system; and an AI-based voice recognition server configured to receive voice data from the relay phone system and transmit the voice recognition result to the relay phone system, wherein the relay phone system extracts only voice data from the voice data packet transmitted from the calling terminal. a voice processing module that transmits; and an STT (speak to text) interworking interface that receives recognized voice recognition information and natural language processed natural language information from the voice recognition server.

Description

A system that can introduce artificial intelligence-based voice recognition service without modifying the existing in-house phone system {A SYSTEM THAT CAN INTRODUCE ARTIFICIAL INTELLIGENCE-BASED VOICE RECOGNITION SERVICE WITHOUT MODIFYING THE EXISTING IN-HOUSE PHONE SYSTEM}

The present invention relates to an artificial intelligence-based voice recognition provision system for providing an artificial intelligence-based voice recognition function to an Internet phone system. Specifically, a system that can introduce an artificial intelligence-based voice recognition service without modifying the existing in-house phone system. It's about.

Recently, the ARS system has been developed and is widely used, but many people complain of frustration with the ARS system or voice recognition system, which is manual and does not communicate two-way, and is often connected to a counselor without entering the correct information, so the system instead Costs increase with deployment, but the waste of counselor manpower remains.

At this time, CTI (Computer Telephony Integration) is used for accurate connection of ARS and smooth consultation distribution in the call center. In this regard, Korean Patent Publication No. 2012-0070417 (published on June 29, 2012) discloses a call center PBX link device to allow general-purpose IPPBXs and CTI middleware from various manufacturers to easily interconnect without changing the protocol. .

However, in order to use CTI more smoothly, a method of adding artificial intelligence-based voice recognition function to the in-house phone system has recently been considered. FIG. 1 is a diagram showing the minimum configuration for performing a voice recognition function in an in-house telephone system 10 equipped with a conventional CTI module 100, as described above.

As shown in FIG. 1, in order to provide an AI-based voice recognition function in the in-house phone system 10 equipped with the CTI module 100, it is necessary to not only build its own AI voice recognition server 30, but also build an AI voice recognition server. A voice reception module capable of communicating with (30) and an STT (Speak-to-text) interface must be built within the in-house telephone system (10).

However, in the case of a company's in-house phone system, for example, a CTI-applied call system or a Centrix or IP-PBX call system configured with VoIP must be modified, but these call systems usually last for a long time after the first call system is introduced and applied. It is common to use it without modification for a period of time, and in order to integrate with a specific system such as AI or voice recognition system, functions for this (voice reception function, STT linkage function) must be developed in the existing service system and applied to the actual service system. do.

However, this is a task that requires a complete modification of the existing in-house telephone system that is in daily service, and although cost is also a problem, in most cases, the telephone system is used for a long period of time without modification or addition of additional functions after introduction. Development is impossible without the development support of the company that developed the system, and in order for the company that developed the telephone system to modify it, it must be separately developed for the specific system to be applied and tailored to its product according to the characteristics of the introduced product. There are difficulties.

The purpose of the present invention is to provide a system that can introduce an artificial intelligence-based voice recognition system without modifying the existing in-house telephone system.

In order to solve the above-described problem, the present invention provides a system for providing an artificial intelligence-based voice recognition service without modifying the in-house telephone system, and the system includes:

In-house telephone systems, including Internet telephone systems;

a relay telephone system configured to receive a call connection request from a calling terminal to the Internet telephone system and perform a call connection to the Internet telephone system; and

It includes an AI-based voice recognition server configured to receive voice data from the relay phone system and transmit the voice recognition result to the relay phone system,

The relay telephone system is,

A voice processing module that extracts and transmits only voice data from voice data packets transmitted from the calling terminal; and

It includes an STT (speak to text) interworking interface that receives recognized voice recognition information and natural language processed natural language information from the voice recognition server.

In the above-described aspect, the voice recognition server

A voice recognition module that converts voice data into voice recognition information in text form through a voice recognition model created through deep learning;

A natural language processing module that extracts numbers and meaningful words from voice recognition information in text form or analyzes sentence relationships; and

It includes an interlocking interface that transmits voice recognition information and natural language processing information from the voice recognition module and natural language processing module to the STT interlocking interface.

In addition, in any of the above-described aspects, the STT interworking interface is configured to receive voice recognition information and natural language processing information from the voice recognition server, and transmit the received voice recognition information and natural language processing information to the in-house telephone system.

In addition, in any of the above-described aspects, the Internet phone system includes a CTI server module that performs functions including extracting phone information, monitoring call status, call forwarding, call creation, and call disconnection, and the CTI server module performs outgoing calls. It is configured to provide information about customers of the terminal.

In addition, in any of the above-described aspects, the in-house telephone system includes a web interface module, and the STT interworking interface receives voice recognition information and natural language processing information from the voice recognition server, and uses the received voice recognition information and natural language processing information. It is transmitted to the web interface module of the in-house phone system, and the web interface module operates to display the received voice recognition information and natural language processing information on the user's monitor.

According to the present invention, a relay phone system capable of relaying voice recognition between an in-house phone system and an AI voice recognition server is provided, making artificial intelligence services including voice recognition functions more convenient without changing the existing in-house phone system. and can be provided effectively.

1 is a diagram showing an example of a conventional system configuration for providing an AI voice recognition service to an in-house telephone system;
Figure 2 is a diagram showing the configuration of a system that provides an AI voice recognition service to an in-house phone system according to an embodiment of the present invention;
Figure 3 is a flowchart showing the processing flow over time among components in the artificial intelligence-based voice recognition providing system according to the present invention.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and will be implemented in various different forms.

The examples herein are provided to make the disclosure of the present invention complete and to fully inform those skilled in the art of the scope of the invention. And the present invention is only defined by the scope of the claims. Accordingly, in some embodiments, well-known components, well-known operations and well-known techniques are not specifically described in order to avoid ambiguous interpretation of the present invention.

Like reference numerals refer to like elements throughout the specification. Also, the terms used (mentioned) in this specification are for describing embodiments and are not intended to limit the present invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. Additionally, components and operations referred to as 'including (or, including)' do not exclude the presence or addition of one or more other components and operations.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings that can be commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless they are defined.

Hereinafter, preferred embodiments of the present invention will be described with reference to the attached drawings. Figure 2 is a block diagram to explain the concept of providing an artificial intelligence-based voice recognition service to an in-house phone system (in-house phone system) 10 that does not provide an existing artificial intelligence-based voice recognition service according to the present invention. .

As shown in Figure 2, the voice recognition providing system 1 includes a relay telephone system 40 that receives an outgoing call from an outgoing terminal; An in-house telephone system (10) linked to a relay telephone system and call transfer; It includes an AI voice recognition server 30 that receives voice data from the relay phone system 40 and transmits the voice recognition result to the relay phone system 40.

In this embodiment, a network is deployed between the calling terminal and the relay telephone system. The network may be a mobile communication network, the Internet such as VoIP (Voice over Internet Protocol) voice packet network, or a telephone network such as PSTN (Public Switched Telephone Network) or PSDN (Packet Switched Data Network).

When the relay phone system 20 receives an outgoing call to the in-house phone system 10, it performs telephone interworking with the in-house phone system 10 and simultaneously transmits voice data to the AI voice recognition server 30 and AI voice recognition. It is operated to receive voice-recognized sentences and/or natural language processed information from the server 30 and transmit them to the in-house telephone system 10.

The relay telephone system 20 includes a voice processing module 22 and an STT interworking interface 24 to perform this operation. In the VoIP system, voice is compressed with a codec and transmitted as RTP (Realtime transfer protocol) packets. The voice processing module 22 decompresses it in real time to enable voice recognition in the voice recognition server 30 and converts it to PCM. It functions to restore the data in its original form and transmit it to the voice recognition server. In addition, RTP voice is transmitted in segments of 10 to 20 ms, and the voice processing module 22 collects it appropriately and performs buffering enough to process voice recognition in the voice recognition server. In the case of a voice recognition server, hundreds of calls must be processed at the same time, so the voice processing module 22 independently performs codec decoding processing for each call in real time, performs buffering processing, and links with the voice recognition server 30 to transmit data. It performs real-time transmission and communication with the interface.

Additionally, the relay telephone system 20 includes an STT interworking interface 24. The voice recognition server 30, which processes STT (Speak To Text), functions to recognize each file or buffer by voice on the local server. The voice recognition server is configured to perform multiprocessing to independently process multiple calls, and the STT interoperability interface 24 performs interfacing with the voice recognition server. In addition, the STT interworking interface 24 also performs the function of transmitting voice recognition information or natural language processed information to the administrator web program (customer management program) or agent web interface module 14. The STT interworking interface 24 is configured to provide a C/S service so that multiple calls can be recognized independently.

The voice recognition server 30 is a centralized voice recognition processing server that recognizes voices independently for each call channel and performs tasks such as number recognition, sentence structure analysis, and entity name recognition through natural language processing of the recognized sentences. It is composed. The voice recognition server 30 can be configured to not only perform voice recognition but also perform other intelligent processing. The voice recognition server 30 may include a voice recognition module 32, a natural language processing module 34, an interworking interface 36, and a voice recognition DB 38 to perform such operations.

The voice recognition module 32 performs the function of converting PCM voice data into text through a voice recognition model created through deep learning. This is also simply called STT (Speech to Text). The information recognized by the voice recognition module 32 is converted into txt format and transmitted to the STT interconnection interface 24 of the relay system 20 through the interconnection interface 36.

The natural language processing module 34 analyzes text sentences recognized by the voice recognition module 32 and performs tasks such as extracting numerical parts from text sentences, extracting words with main meaning, or analyzing sentence relationships. And the sentences or information analyzed by the natural language processing module 34 are transmitted to the relay phone system through the interworking interface 36.

The interworking interface 36 receives decoded voice data in conjunction with the voice processing module 22, such as a phone system using VoIP or an AI speaker, and transmits it to the voice recognition module 32, and conversely, the voice recognition module 32 And/or performs the function of transmitting sentences and information recognized from the natural language processing module 34 back to the telephone system and AI speaker. In addition, since each call has separate call information (Call ID), it also performs the function of updating and storing this information.

The in-house telephone system 10 represents a telephone system installed in an existing in-house phone system. In the embodiment shown in FIG. 2, the in-house telephone system 10 includes an Internet telephone system 110 and a customer management program 120.

The Internet phone system 110 of the in-house phone system 10 is interconnected with the relay phone system 20, and the Internet phone system 110 includes a CTI server module 112. The CTI server module 112 performs various functions such as extracting information of a call connected to a call through the relay telephone system 20, monitoring and transmitting call status, and call-related auxiliary functions, such as call forwarding, call creation, and call disconnection. When used in systems such as CRM, it is used to construct a system that expands the other party's information by linking with the DB by transferring information and monitoring call status.

The customer management program 120 refers to a program configured to use information transmitted from the CTI server module 112 in conjunction with systems such as CRM and ERP to use customer information such as customer status, previous history, and call progress. .

In this embodiment, the customer management program 120 further includes a web interface module 124 to provide information recognized by the voice recognition server 30 to the in-house telephone system 10. The web interface module 124 is configured to provide sentences and/or natural language processed information recognized from the STT linking interface 24 in the form of a pop-up window on the user's (counselor) screen.

Here, the web interface module 124 may be run in a customer management program so that recognized sentences and/or natural language processed information are provided to the agent's monitor using a script executed in a web browser. Here, a web browser is a program that allows the use of web (WWW: world wide web) services and refers to a program that receives and displays hypertext written in HTML (hyper text mark-up language). For example, Netscape. , Explorer, Chrome, etc.

Figure 3 is a flowchart showing the operation flow in the voice recognition providing system as described above. As shown in Figure 3, the caller's terminal requests a call connection to the in-house telephone system 10. The relay phone system 20 is configured to interoperate with the in-house phone system 10, so that the relay phone system 20 receives a call connection request. Therefore, when the relay phone system 20 receives a call connection request, the relay phone system performs a call connection to the in-house phone system 10.

After call interconnection is established between the relay phone system (20) and the in-house phone system (10), voice data from the caller's terminal is transmitted to the relay phone system (20), and the relay phone system transmits voice data to the in-house phone system (10). At the same time as transmitting data, the voice processing module 22 of the relay telephone system 20 decompresses the voice data and transmits the decompressed voice data to the voice recognition server 30 to request voice recognition.

Since the caller's terminal and the in-house phone system (10) are connected through the relay phone system (20), the caller makes a call with a counselor of the in-house phone system (10), and at the same time, the voice recognition server (30) recognizes voice data buffered to a certain size from the voice processing module 22, converts it into text form, and performs natural language processing on the voice-recognized data to extract necessary information.

The voice-recognized text sentences and natural language processed information generated from the voice recognition server 30 are transmitted to the relay phone system 20 through an established interworking interface between the voice recognition server and the relay phone system, and the relay phone system 20 )'s STT interworking interface 24 transmits voice-recognized text sentences and natural language processed information to the customer management program 120 of the in-house phone system.

As described above, in the embodiment of the present invention, the number coming into the in-house Internet phone system is connected to the phone relay system to send a call, and the phone relay system sends the call back to the original call system, the Internet phone system. . In this process, the telephone relay system processes call information and voice data, transmits them to the voice recognition server, receives voice recognition results from the voice recognition server, and transmits them to the customer management program included in the in-house phone system.

In the above-described embodiment, this was explained using the Internet phone system as an example. This is because the use of voice phone systems using Internet communication such as IP-PBX or Centrix is common these days, so if developed in conjunction with the Internet phone system, it is possible to link with most call networks, and it is also possible to use CTI, etc., rather than analog or Internet phones. Even if it is a PBX system, linking with the Internet phone network is relatively simple, so it can be applied to most in-house phone systems.

In addition, in the case of customer consultation programs used by final-stage counselors, in most cases, the service company continues to develop and manage additional services, so linking the information processed by the telephone relay system to the customer consultation program and delivering it can be developed without any major problems. possible. Usually, these customer consultation programs are made as web programs or C/S programs, and can be implemented through REST API, JSON linkage, message linkage, etc.

According to this embodiment of the present invention, a relay phone system (20) capable of relaying voice recognition between an in-house phone system and an AI voice recognition server is provided, thereby providing an artificial intelligence device including a voice recognition function without changing the existing in-house phone system. Intelligent services can be provided more conveniently.

The device described above may be implemented with hardware components, software components, and/or the device described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in the embodiments include, for example, a processor, a controller, an Arithmetic Logic Unit (ALU), a Digital Signal Processor, a microcomputer, and a Field Programmable Gate Array (FPGA). , may be implemented using one or more general-purpose computers or special-purpose computers, such as a Programmable Logic Unit (PLU), a microprocessor, or any other device that can execute and respond to commands. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, there are cases where a single processing device is described, but those skilled in the art will understand that the processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.

Software may include a computer program, code, commands, or a combination of one or more of these, and may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. It can be embodied in . Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (Magneto-Optical Media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

As described above, although the embodiments have been described with limited examples and drawings, various modifications and variations can be made by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, the scope of protection of the present invention should be interpreted in accordance with the claims below rather than being limited by the above-described embodiments, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of rights of the present invention.

10: In-house phone system 20: Relay phone system
22: Voice processing module 24: STT interlocking interface
30: AI-based voice recognition server 32: Voice recognition module
34: Natural language processing module 36: Link interface
38: Voice recognition DB 110: Internet phone system
112: CTI server module 120: Customer management program
122: CTI client module 124: Web interface module

Claims (5)

In a voice recognition provision system that allows the introduction of an artificial intelligence-based voice recognition service without modifying the in-house phone system,
In-house telephone systems, including Internet telephone systems;
a relay telephone system connected to the in-house telephone system through a network independent of the in-house telephone system to receive a call connection request from a calling terminal to the Internet telephone system and perform a call connection to the Internet telephone system; and
It includes an AI-based voice recognition server configured to receive voice data from the relay phone system and transmit the voice recognition result to the relay phone system,
The relay phone system is,
A voice processing module that extracts and transmits only voice data from voice data packets transmitted from the calling terminal; and
It includes an STT (speak to text) interworking interface that receives recognized voice recognition information and natural language processed natural language information from the voice recognition server,
The STT interworking interface is configured to receive voice recognition information and natural language processing information from the voice recognition server, and transmit the received voice recognition information and natural language processing information to the in-house telephone system,
The in-house phone system includes a web interface module,
The STT linkage interface receives voice recognition information and natural language processing information from the voice recognition server, and transmits the received voice recognition information and natural language processing information to the web interface module of the in-house phone system,
The web interface module operates to display received voice recognition information and natural language processing information on the user's monitor,
After call interconnection is established between the relay phone system and the in-house phone system, voice data from the caller's terminal is transmitted to the relay phone system, and the relay phone system transmits the voice data to the in-house phone system and at the same time, the voice data of the relay phone system is transmitted. The processing module decompresses voice data, transmits the decompressed voice data to the voice recognition server, and requests voice recognition.
The voice recognition server is configured to collect decompressed voice data and transmit voice recognition information to the relay phone system each time a sentence is completed,
The STT interworking interface of the relay phone system is configured to transmit voice-recognized text sentences and natural language processed information to the customer management program of the in-house phone system.
Voice recognition system.
According to paragraph 1,
The voice recognition server is
A voice recognition module that converts voice data into voice recognition information in text form through a voice recognition model created through deep learning;
A natural language processing module that extracts numbers and meaningful words from voice recognition information in text form or analyzes sentence relationships; and
Characterized in that it includes an interlocking interface that transmits voice recognition information and natural language processing information from the voice recognition module and natural language processing module to the STT interlocking interface.
Voice recognition system.
delete According to paragraph 1,
The Internet phone system includes a CTI server module that performs functions including phone information extraction, call status monitoring, call forwarding, call creation, and call termination, and the CTI server module is configured to provide customer information of the calling terminal. characterized by
Voice recognition system.
delete