KR100929531B1 - Information provision system and method in wireless environment using speech recognition - Google Patents

Abstract

The present invention relates to a system and method for providing information in a wireless environment using speech recognition.

The information providing system is connected to at least one user terminal through a network, and provides information according to a request from the user terminal. The voice signal for requesting the user's information is processed into streaming voice data in the form of voice recognition processing and then provided to the system through the network. The system detects an endpoint for the received streaming voice data and performs a speech recognition process to recognize what information the user has requested. The information according to the recognition result is processed into data in a form that can be transmitted to the user terminal.

Accordingly, by preprocessing and transmitting the voice signal of the user inputted to enable the voice recognition processing in the user terminal, it is possible to prevent distortion from occurring during voice data transmission.

Description

Information Offering System and Method using Voice Recognition in Wireless Environment

1 is a network connection diagram of an information providing system according to an exemplary embodiment of the present invention.

2 is a structural diagram of a user terminal according to an exemplary embodiment of the present invention.

3 is a detailed structural diagram of an information providing system according to an exemplary embodiment of the present invention.

4 and 5 are flowcharts of an information providing method according to an exemplary embodiment of the present invention.

6 is a flowchart illustrating a processing process based on a voice recognition processing result according to an exemplary embodiment of the present invention.

7 and 8 are exemplary diagrams of information output to a user terminal according to an information providing method according to an exemplary embodiment of the present invention.

The present invention relates to an information providing system, and more particularly, to an information providing system, a method and an information display apparatus using speech recognition in a wireless environment.

In general, a system for providing predetermined information to users through a telephone is a customer consultation system that provides a plurality of information in response to a user's request when the user connects through wired or wireless communication. Conventional customer consultation system is mainly made in the form of connecting the customer with a predetermined counselor when the customer is connected by wire or wireless, so that the customer can receive the desired information by a direct call between the customer and the counselor. In another form, when the customer communicates by wire or wirelessly, the customer is connected to an ARS (automatic response system) server so that the customer can receive information through the ARS server. In another form, when a customer requests a menu of a desired service by voice on a terminal, the voice is analyzed to provide corresponding information.

A system for providing information using speech recognition provides a service based on a telephone network, or provides information based on a wireless network, which is divided into a first form and a second form according to a speech processing method.

In the first aspect, when the terminal merely transmits an analog voice file in which a user's voice is recorded to the system, the system processes the voice file according to a predetermined voice recognition process to generate voice data, thereby performing voice recognition. For example, the terminal transmits the analog voice file to the system through the telephone network, and the voice recognition engine of the system processes and recognizes the analog voice file transmitted through the exchange and the interactive voice response (IVR) server according to a predetermined voice recognition process. do. In this case, however, the terminal transmits an analog voice file to the system, which is accompanied by distortion and noise.

On the other hand, the second form is a method in which the terminal transmits data obtained after the predetermined voice recognition processing of the voice file, rather than transmitting an analog voice file recording the user's voice. This method is called distributed speech recognition (DSR), and the terminal performs a major part of the general speech recognition process based on a speech file.

However, in this case, since the process of the terminal is complicated, the load of the terminal is increased, and as the hardware for implementing the process is required, the terminal manufacturing price increases.

Therefore, the technical problem to be achieved by the present invention is to solve the above-described problems, and to provide information corresponding to the terminal of the user by performing the speech recognition more efficiently in a wireless environment.

In addition, the technical problem to be achieved by the present invention is to reduce the processing load of the terminal to perform the speech recognition without distortion, so that accurate information is provided to the terminal.

Another object of the present invention is to display a voice recognition state on a terminal so that a user can easily check the same.

The information providing method according to the characteristics of the present invention for achieving the technical problem of the present invention, in the system is connected to at least one user terminal and the counselor terminal through a network, providing information in response to a request from the user terminal A method for converting and compressing an analog voice of a user into a digital signal as it is using a separately configured codec while requesting information from the system by the user terminal, and processing the information into a predetermined format that can be transmitted through the wireless network. Transmitting to the system as post streaming voice data; b) if there is a request for providing information through the user terminal, the system receiving streaming voice data transmitted from the user terminal; c) restoring the streaming voice data by the system to generate voice data; d) the system performing endpoint detection on the speech data; e) the system performing a speech recognition process on the speech data based on the detected endpoint; And f) finding and transmitting the corresponding information to the user terminal according to the speech recognition result.

In addition, the information providing system according to another aspect of the present invention is a system for providing information in response to a request from the user terminal, connected to at least one or more user terminals and counselor terminals through a network, requesting information provision, and separately A user terminal which converts and compresses an analog voice of a user into a digital signal using a configured codec, processes the same into a predetermined format that can be transmitted through the wireless network, and then transmits the streaming voice data; In response to a request for providing information from the user terminal, the analog voice of the user is converted into a digital signal as it is, compressed, and then received and processed in a predetermined format that can be transmitted through the wireless network. A voice processing unit; An endpoint detection unit for performing endpoint detection on the generated voice data; A speech recognition engine that performs a speech recognition process on the speech data based on the detected endpoint; And a service control unit for searching for and transmitting the corresponding information to the user terminal according to the voice recognition result.

In this case, the user terminal may include an interface unit for inputting a user voice signal; A voice signal processor configured to process the user voice signal input through the interface unit as streaming voice data through a separately installed voice codec; A transmitter / receiver for transmitting the streaming voice data to the system through the network and receiving a signal transmitted from the system; And a display unit for displaying the processing information including the information transmitted from the system as a result of the voice recognition processing transmitted from the system. In particular, the voice signal processor converts and compresses an analog voice signal of a user into a digital signal, and then generates and transmits the transmissionable streaming voice data of a predetermined format to the system.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention.

1 is a diagram illustrating a network connection of an information providing system according to an embodiment of the present invention. As shown in FIG. 1, the information providing system 100 according to an embodiment of the present invention may be provided through a wireless network 200. It is connected with a plurality of user terminals 300.

Here, the user terminal 300 is a communication device capable of connecting to the information providing system 100 via a wireless network 200 (also referred to as "network" for convenience of description below), for example, a mobile communication terminal, Communication devices capable of all wireless network connections, such as Internet phones and PDAs, may be included.

2 is a structural diagram of a terminal according to an embodiment of the present invention.

As shown in FIG. 2, the user terminal 300 according to an exemplary embodiment of the present invention may include an interface unit 310, a voice signal processor 320 for processing a voice signal input through the interface unit 310, and a processed voice signal. And a transmission / reception unit 330 for transmitting the data to the system 100 through the network 200, and a display unit 340 for displaying various types of information such as a result of voice recognition processing and information transmitted from the system. In addition, the character information processing unit 350 to process the information in the form of the character to display on the display unit 340, the image information processing unit 360 to process the information in the form of the image displayed on the display unit 340, and to transmit the audio signal The service control unit 370 further controls reception and processing of information including text and images.

The interface unit 310 performs a function for interfacing with a user, and includes, for example, a microphone for converting and outputting a user's voice into an electrical voice signal and a speaker for outputting a voice signal. It includes an input device.

The voice signal processor 320 processes voice signals of the user input through the interface unit 310 and generates predetermined voice data to be sent to the system. In detail, the voice signal processor 320 converts an analog voice signal input through the interface unit into a digital voice signal, and in particular, converts the analog voice signal into a digital voice signal according to a set bit rate. The digital voice signal is compressed and processed into streaming data, and then transmitted to the system 100 through the transceiver 340. In particular, the voice signal processor 320 records an input user voice signal, and if the voice recognition fails, transmits a voice recording file to the system 100 as necessary. In addition, the voice signal processor 320 processes the voice signal transmitted from the system 100 to be output through the interface unit 310. Hereinafter, the voice signal output from the user terminal and transmitted to the system will be referred to as streaming voice data. Meanwhile, the voice signal processor 320 may be called a voice codec, and the text information processor 350 may be a text codec, and the image information processor 360 may be called a video codec.

The transmitter / receiver 340 processes streaming voice data, which is a digital voice signal of a predetermined format, provided from the voice signal processor 320 as a transmittable signal and transmits the signal to the system 100, and also transmits various forms transmitted from the system 100. Receive a signal containing information.

The text information processor 350 extracts the text information from the signal received from the transceiver 340, and then processes the text information into a form that can be displayed on the display. The video information processor 360 extracts the video information from the received signal. Process in a form that can be displayed. The image information includes both a still image and a video.

The service controller 370 controls the above components 310 to 360 so that a service for providing information through voice recognition according to an embodiment of the present invention is provided through the user terminal. For example, operating based on an application for providing information through voice for each service menu, the information corresponding to the menu requested by the user according to the link between the terminal 300 and the system 100 includes voice. It can be provided to the user in various forms.

In addition to the components described above, the user terminal 300 according to an embodiment of the present invention may further include other components for performing terminal-specific functions (for example, a call processing function including call connection processing). have.

Meanwhile, the information providing system 100 for transmitting predetermined information to the terminal in response to a request from the terminal 300 as described above has the following structure.

3 is a detailed structural diagram of an information providing system according to an exemplary embodiment of the present invention.

Information providing system 100 according to an embodiment of the present invention, as shown in Figure 1 and 3, includes a voice recognition server 10, in addition to the agent management server 20. In addition, the call processing unit 50 for the call connection with the user terminal may be further included.

The information providing system 100 may be connected through a mobile communication company's FEP server (not shown). When the FEP server requests information from the user terminal, the FEP server performs a function such as a voice recognition server for providing the information and the location of the base station, so that the efficient connection between the information providing system 100 and the user terminal 300 can be made. have. Of course, other servers that perform the same function may be used in addition to the FEP server.

The voice recognition server 10 processes a voice signal of a predetermined format transmitted from the user terminal 300 to perform voice recognition. To this end, the voice processing unit 11 which restores the specifically input streaming data and generates corresponding voice data, and performs end-point detection (hereinafter referred to as EPD) processing on the restored voice data. The speech recognition engine 13 which performs speech recognition on the speech data based on the detected end point detected by the EPD processor, and generates text information and speech recognition result corresponding to the speech data, and the text. The service control unit 14 provides information and a voice recognition result to the counselor management server 20 or provides the user terminal 300 with information retrieved according to the voice recognition result and the voice recognition. Meanwhile, the service controller 14 may perform communication control for synchronization between the user terminal 300 and the system 100.

The speech recognition engine 13 according to an exemplary embodiment of the present invention processes the speech recognition result into a predetermined form of information and transmits the speech recognition result to the user terminal 300 so that the user can check the speech recognition result of the system. In this case, the voice recognition result transmitted to the user terminal may be in the form of a text or may be complex media information such as an image or a voice. In addition, the speech recognition engine 13 may include a speech recognition DB (not shown) in which basic information for speech recognition of the input speech data is stored. The voice recognition DB may store a list of recently used or frequently used words for each customer or a list of frequently used words for all customers.

Information requested by a user is provided based on a voice recognition result of the voice recognition server 10, and a plurality of agent 1 group terminals 30 and agent 2 group terminals 40 may be connected to the agent management server 20. . The agent 1 group terminal 30 is connected to the agent management server 20 and the voice recognition engine 13, and the agent 2 group terminal 40 is connected to the agent management server 20 and the call processing unit 50. .

The agent management server 20 manages the agent 1 group terminal 30 and the agent 2 group terminal 40, and in particular, selects a predetermined agent terminal according to a user's information request so that corresponding information is provided to the user terminal. .

In particular, when the voice recognition result of the voice recognition engine 13 is smaller than a preset reference value, the agent management server 20 selects one of the agent 1 group terminals 30, and the voice recording file transmitted from the user terminal. Character information including a list of recognized words recognized by the speech recognition engine 12 is provided to the selected counselor terminal. As a result, a corresponding counselor recognizes a user request based on the voice recording file reproduced through the selected counselor 1 group terminal 30 and the text information, and transmits the recognized result to the speech recognition engine 13. The voice recording file is a file transmitted from the user terminal.

On the other hand, the agent management server 20 is one of the agent 2 group terminals so that when the voice recognition by the agent 1 group through the agent 1 group terminal 30 can be heard directly by the agent of the agent 2 group ( Select 40). Thereafter, the counselors of the selected two group terminals directly listen to the user's voice and transmit the corresponding processing result to the call processor 50. At this time, the counselors of the two counselors listen directly to the voice of the user instead of directly talking to the user, and the response to the user may be performed by the call processor 50.

The call processing unit 50 is a server that performs call processing according to a call connection request from the user terminal 300, as shown in FIG. 3, and includes an internal exchange 51 connected to an external exchange of a wireless communication company or a wire communication company, It includes a call distribution server (CTI: computer and telephony integration, 52) for distributing a plurality of calls, and an interactive voice response (IVR) server 53.

The internal exchange 51 is directly connected to the wired / wireless communication terminal possessed by the customer through the external exchange of the wired / wireless communication company, and the customer provides the information providing service through voice recognition according to an embodiment of the present invention. To receive it. The call distribution server 52 is connected to the exchange 51 and provides the registered information by forming a network with the information and the established information and control of the connected devices as well as the sharing of information resources between the telephone and the computer. In particular, when a call connection request is input from the user terminal 300, the call distribution server 52 connects the user terminal 300 to one IVR server among a plurality of IVR servers, and the IVR server 53 provides a predetermined scenario service. Providing a voice announcement according to the user terminal 300 so that the user requests the desired information. To this end, the IVR server 53 stores and manages voice information data obtained by voicening a plurality of announcements and information to be provided. When the call with the user terminal is connected, the IVR server 53 reproduces and outputs the stored announcements. In addition, information corresponding to a user request provided by the counselor terminal 40 may be provided as a voice or processed as complex media data such as a text or an image and provided to the user terminal 300. In particular, in the embodiment of the present invention, the call processing unit 50 allows a call connection to be made between the agent 2 group terminal 40 and the user terminal 300 when the voice recognition for the user request fails. In this case, the counselor belonging to the counselor 2 group can directly connect with the user through the counselor 2 group terminal 40 to directly listen to the information requested by the user.

On the other hand, the information providing system according to an embodiment of the present invention may further include a text-to-speech (TTS) server (not shown) for converting text data into voice.

In the present embodiment, although components of each server are shown to operate in the corresponding server, the present invention is not limited thereto and may be implemented as independent servers to process corresponding functions. In addition, each server and the components constituting the server are classified according to their functions, and are not limited to those classified as described above. For example, the voice management server may perform a function of the agent management server.

Next, an information providing method according to an exemplary embodiment of the present invention will be described based on the structure.

4 and 5 are flowcharts of an information providing method according to an exemplary embodiment of the present invention.

The user connects the terminal 300 to a predetermined kit or a hot key to receive the predetermined information from the terminal 300 through the information providing system 100 according to an exemplary embodiment of the present invention. When pressing to request a network connection, as shown in FIGS. 4 and 5, the service controller 370 of the terminal 300 connects to the information providing system 100 on the network 200 through the transceiver 340. Try. When the connection is made, a service menu may be transmitted from the information providing system 100 to be displayed on the terminal 300 (S100). In this case, the service menu is stored in the terminal and may be displayed when a connection to the system is made. If necessary, the service menu may be displayed after the menu update is made through the system.

Thereafter, when the user selects a predetermined menu, the service controller 370 of the terminal 300 requests a service while transmitting information corresponding to the selected menu to the information providing system 100 (S110). In response to the service request, the information providing system 100 performs authentication on the user terminal 300 (S120). The authentication of the terminal may determine whether the user is registered as a member capable of providing a service based on the identification number of the terminal (eg, a unique number given when the terminal is manufactured). This authentication process may optionally be performed.

In response to the service request, as shown in FIG. 5, the information providing system 100 requests the terminal 300 to transmit user voice data according to the selected service menu. Then, the terminal 300 processes the voice of the user input through the interface unit 310 and transmits it to the system 100 (S130 to S140). That is, the voice signal processor 320 of the terminal 300 processes the voice signal of the user input through the interface unit 310 into streaming voice data of a predetermined format, and then transmits the voice signal to the system 100 through the transceiver 340. To pass. At this time, the voice signal processor 320 records the input user voice signal and stores it as a predetermined file.

Through the above process, streaming voice data transmission is performed between the terminal 300 and the system 100. As such, the voice data transmitted from the user terminal 300 to the system 100 is transmitted in a state in which a predetermined preprocessing is performed to enable the voice recognition process in the system 100 instead of the analog voice signal itself. Distortion can be prevented from occurring. In addition, the terminal includes only means for implementing the preprocessing, thereby preventing problems such as an increase in cost and complexity of the structure according to the implementation of the means.

Meanwhile, the streaming voice data transmitted from the user terminal 300 is transferred to the voice recognition server 10 (S150), and the voice recognition server 10 performs voice recognition on the delivered streaming voice data. In detail, as shown in FIG. 4, the corresponding streaming data is restored (decoded, etc.) to generate corresponding voice data (S160), and a voice section, that is, a start point and an end point of the voice is extracted from the restored voice data. In operation S170, voice recognition is performed on the voice data based on the extracted endpoint. At this time, the speech recognition engine 12 of the speech recognition server 10 may perform speech recognition by searching a speech recognition DB according to a predetermined scenario.

When the recognition score, which is one of the speech recognition results, is equal to or greater than a predetermined reference value, the service controller 15 completes the information search using the speech recognition result (S29), and provides the searched information to the user terminal 300 ( S190 to S200). In this case, the retrieved information may be provided to the user as various information such as text, voice, graphic, or a complex form of text and graphic. As described above, data transmission and reception between the information providing system 100 and the user terminal 300 is not performed using a voice network through an exchange, but data is transmitted and received such as a TCP / IP network. By performing through a network, more accurate information can be provided according to voice recognition.

Meanwhile, the information provided by the system 100 may be differently performed according to the result of the speech recognition process.

6 is a flowchart illustrating a processing process based on a voice recognition processing result according to an exemplary embodiment of the present invention.

If the voice recognition result is successful, as described above, information according to the recognition result is provided to the user terminal 300 through the data network (S300 to S320). However, if the speech recognition result by the speech recognition engine 12 is smaller than the preset reference value, the service controller 15 notifies the user terminal 300 of the speech recognition failure so that the user responds again and provides new speech data. Receive (S330).

At this time, a text or voice indicating a failure in speech recognition is displayed or output through the user terminal 300 to confirm the voice recognition state, and then retry the information request through the voice input. At this time, the voice signal processor 320 of the user terminal 300 processes and transmits the input voice signal as streaming voice data according to the retry, and transmits the voice recording file recording the voice signal. Then, the voice recognition server 10 of the system 100 performs voice recognition processing of the retransmitted streaming voice data (S340 to S350).

Thereafter, the service controller 15 of the system 100 selects one of the counselor 1 group terminals 30 through the agent management server 20, and recognizes the list of recognized words recognized by the speech recognition engine 12 with the selected agent terminal. By providing a voice recording file transmitted from the user terminal 300 with the text information including the, so that the recognition by the counselor 1 group can be performed (SS360). On the other hand, the service control unit 15 may call the counselor group 1 as described above only when the voice recognition result by the received voice data is smaller than the preset reference value according to the re-response.

The counselor group 1 terminal 30 reproduces and outputs the voice recording file while displaying a list of recognized words transmitted from the voice recognition engine 40. Therefore, the counselor belonging to the counselor group 1 recognizes the user's voice by listening to the recording file through the headset. If a result of the recognition is found in the list of recognized words, the selected word is selected. If not, the predetermined word is searched and the search result word is entered. The counselor 1 group terminal 30 transmits the result to the service controller 15 through the agent management server 20 when a result word recognized from a counselor belonging to the counselor 1 group is selected or input. Thereafter, the service controller 15 determines that the voice recognition is successfully completed by the counselor group 1, performs information search using the recognized result, and provides the searched information to the user terminal 300 (S370 to S380). At this time, the agent belonging to the first group of agents, not the service control unit 15, may perform information search based on the directly recognized result word. In this case, the service control unit 15 may transmit the retrieved information to the terminal 300. Only can be done.

On the other hand, as a result of the recognition by the first group of agents in step S370, when the agent belonging to the first group of agents listens to the recording file, the customer's voice cannot be judged due to unclear pronunciation of the customer, ambient noise, etc. In the case of failure due to the cause, the service control unit 15 directly connects the terminal of the corresponding user to the predetermined agent 2 group terminal 40 selected by the agent management server 20 through the call processing unit 50. (S390). Accordingly, the counselor belonging to the counselor 2 group can directly connect with the user through the counselor 2 group terminal 40 to directly listen to the information requested by the user. At this time, the agent belonging to the agent 2 group does not directly talk to the user, but the IVR server 53 of the call processing unit 50 requests the voice input to the user again according to the service scenario caused by the recognition failure by the agent 1 group. Send a message, and as a result, the customer directly inputs the voice through the counselor 2 group terminal 40 to recognize it. That is, since the agent 2 group terminal 40 hears the voice input directly from the user to the agent belonging to the agent 2 group through a headset connected to the phone, the agent belonging to the agent 2 group is directly connected to the user without a direct call with the user. I can hear the voice. At this time, the user terminal 300 transmits the user's voice signal to the system 100 as in the case of a voice call over a general wireless network, and in this case, the terminal 300 and the system 100 through the voice network. Signal transmission is made.

 Therefore, the counselor belonging to the counselor 2 group listens directly to the user's voice through a headset or the like, searches for an information DB (not shown), and inputs the searched result. Thereafter, when the agent 2 group terminal 40 receives a result recognized from an agent belonging to the agent 2 group, the result is transmitted to the IVR server 53, and the IVR server 53 successfully recognizes the voice by the agent 2 group. It is determined that it is completed (S27), performs an information search and provides the searched information to the user through the exchange 10 (S400).

Next, an example of how the information providing method according to an exemplary embodiment of the present invention performed as described above is actually applied will be described.

7 and 8 are diagrams illustrating a process of applying the information providing method according to an embodiment of the present invention. That is, FIG. 7 is a screen that is displayed on a user terminal in connection with an information providing system according to an embodiment of the present invention when a user, ie, a customer, wants to be provided with information on a route to a predetermined destination, for example, COEX, in Seoul. It is an exemplary view showing these.

As illustrated in FIG. 7, when the user selects a menu of “Find Voice” from the <Initial Menu Screen> provided from the system 100, the system 100 requests transmission of voice data to the terminal 300. . Accordingly, the <voice input screen> is displayed on the terminal 300, and the user inputs a destination name to be searched by voice.

When the user inputs the destination name by voice, the terminal 300 processes the input voice signal to generate streaming voice data, and the corresponding destination name is displayed on the screen as shown in FIG. So that the user can check it.

Thereafter, the terminal 300 transmits the streaming voice data corresponding to the destination name <Seoul, COEX>, which has been confirmed by the user, to the system 100, and the system 100 restores, detects the endpoint, and voices as described above. A recognition process is performed to recognize what the destination name entered by the user is. In this case, when the voice recognition is successfully performed or the voice recognition fails and the next voice recognition connected to the agent 1 group or the agent 2 group is performed, the information corresponding to the result of the voice recognition as shown in the <Introduction of the road guide> of FIG. Is provided to the terminal 300 and displayed, so that the user can check. In contrast, the transmitted information may be processed as voice data, and may be output as voice, for example, "CO X is 50m from Samsung Station toward Youngdong Bridge. The telephone number is 02-XXXX-OOOO."

On the other hand, if the voice recognition fails, as shown in <Recognition Failure Screen> of FIG. 8, the information "Failed Voice Recognition" and "Sorry, please try again" is output or the announcement is output. You will check the status. Thereafter, information corresponding to the destination name or the service menu name that the customer responses is provided through the same process as the above scenario.

Such information provision according to the present invention is not limited to providing information in a wireless environment, and various modifications and changes can be made without departing from the technical gist of the invention.

Although the present invention has been described with reference to the most practical and preferred embodiments, the present invention is not limited to the above-described embodiments, and includes various modifications and equivalents falling within the scope of the following claims.

As described above, according to an embodiment of the present invention, when providing information using voice recognition, the user terminal pre-processes and transmits a voice signal input to enable voice recognition processing, thereby transmitting distortion in voice data transmission. Can be prevented from occurring. Therefore, it is possible to perform the speech recognition more efficiently in the wireless environment to accurately provide the requested information to the user's terminal.

In addition, the user terminal includes only means for implementing the preprocessing, thereby preventing problems such as an increase in cost and complexity of the structure due to the implementation of the means. As a result, voice recognition through the system is performed without distortion while reducing the processing load of the terminal.

In addition, by displaying the voice recognition status of the system on the terminal, the user can easily check the voice recognition status.

Claims (13)

In a system connected to at least one user terminal and an agent terminal through a wireless network, in a method for providing information in response to a request from the user terminal, a) The user terminal requests the system to provide the information, converts and compresses the analog voice of the user into a digital signal using a separately configured codec, processes it into a predetermined format that can be transmitted through the wireless network, and then streams the voice data. Transmitting to the system via a network; b) if there is a request for providing information through the user terminal, the system receiving streaming voice data transmitted from the user terminal; c) restoring the streaming voice data by the system to generate voice data; d) the system performing endpoint detection on the speech data; e) the system performing a speech recognition process on the speech data based on the detected endpoint; And f) finding and transmitting the corresponding information to the user terminal according to the speech recognition result; Information providing method comprising a. delete The method of claim 1 The counselor terminal is divided into 1 group terminal and 2 group terminal, Step f) Transmitting, by the system, text information including a voice recording file transmitted from the user terminal and a recognition word list according to a voice recognition result to one group terminal when the voice recognition processing result is smaller than a preset reference value; And Finding information based on a result of the recognition performed based on the text information and the voice recording file in the first group terminal and transmitting the information to the user terminal; Information providing method further comprising. The method of claim 3, If the system fails recognition by the first group terminal, the method further comprises the step of connecting the second group terminal to the user terminal by call, The two-group terminal recognizes the user's request based on the voice signal provided from the user terminal and provides the information to the user terminal. The method of claim 3, And if the voice recognition processing result is successful, processing the voice recognition processing result into at least one of text, voice, video, and composite media information and providing the processed result to the user terminal. The method of claim 1 Streaming voice data between the user terminal and the system is transmitted and received over a data network on the network. In a system connected to at least one user terminal and a counselor terminal through a wireless network to provide information in response to a request from the user terminal, A user terminal for requesting information provision, converting and compressing an analog voice of a user into a digital signal as it is using a separately configured codec, processing it in a predetermined format that can be transmitted through the wireless network, and then transmitting the streaming voice data; In response to a request for providing information from the user terminal, the analog voice of the user is converted into a digital signal as it is, compressed, and then received and processed in a predetermined format that can be transmitted through the wireless network. A voice processing unit; An endpoint detection unit for performing endpoint detection on the generated voice data; A speech recognition engine that performs a speech recognition process on the speech data based on the detected endpoint; And The service control unit for finding the corresponding information according to the speech recognition result and transmits it to the user terminal Information providing system comprising a. The method of claim 7, wherein The counselor terminal is divided into 1 group terminal and 2 group terminal, And a counselor management server for selecting a terminal to respond to a request for providing information from the user terminal among the group 1 or group 2 terminals. The method of claim 8, When the voice recognition processing result is smaller than a preset reference value, the selected one group terminal selects an agent based on text information including a voice recording file transmitted from the user terminal and a recognition word list according to the voice recognition result. Information providing system for performing voice recognition through. The method of claim 9, The information providing system further comprises a call processing unit which connects the one group of two terminals with the user terminal when the recognition fails even by the first group terminal. The method of claim 10, The call processing unit A call distribution server for connecting a call between the user terminal and the group 2 terminal; And A response processing server for transmitting a voice signal provided from the user terminal to which the call is connected to the second group terminal, and providing the response information provided from the second group terminal to the user terminal; Information providing system comprising a. The method according to any one of claims 7 to 11. The user terminal An interface unit for inputting a user voice signal; A voice signal processor configured to process the user voice signal input through the interface unit as streaming voice data through a separately installed voice codec; A transmitter / receiver for transmitting the streaming voice data to the system through the network and receiving a signal transmitted from the system; And A display unit for displaying the processing information including the information transmitted from the system as a result of the voice recognition processing transmitted from the system; Information providing system comprising a. delete

Images