KR20060089288A - Distribute speech recognition system - Google Patents

Abstract

The present invention relates to an information providing system of a telematics system. In particular, the present invention relates to a telematics system using a distributed speech recognition system capable of simultaneously performing a remote speech recognition and a local speech recognition at a remote location and selecting a result that is quickly or accurately recognized by a user. An information providing system and method for a telematics system using a distributed speech recognition system for receiving information provided by a service center of a system. The present invention detects a voice feature of the voice from the spoken voice word to generate voice feature data, generates and transmits the voice feature data, and transmits the information about the voice word in response to the transmission of the voice feature data. A telematics terminal for receiving and notifying a user, a distributed speech recognition server for receiving the speech feature data and performing speech recognition using the speech feature data to transmit a speech recognition recognition word, and receiving the recognition word A database server for finding and transmitting information corresponding to a recognized word, and receiving voice feature data from the telematics terminal and transmitting the received voice feature data to the distributed voice recognition server, and inputting the recognized word to the database server in response to the transmission of the voice feature data. In response to the recognition word transmission. When information is received from the database server for the recognized word comprises a Web server that is transmitted to the telematics terminal.

Telematics, Speech Recognition, DSR, Web Server

Description

System and method for providing information of telematics system using distributed speech recognition system {DISTRIBUTE SPEECH RECOGNITION SYSTEM}             

1 is a diagram illustrating a configuration of a distributed speech recognition system of a telematics system according to the present invention.

2 is a flowchart illustrating a distributed speech recognition method of a telematics system according to a first embodiment for requesting information of the present invention;

3 is a flowchart illustrating a distributed speech recognition method of a telematics system according to a second embodiment for requesting information of the present invention;

4 is a diagram showing the configuration of a telematics terminal according to the present invention;

5 is a flowchart illustrating a first distributed speech recognition method in a telematics terminal according to a second embodiment of the present invention.

6 is a flowchart illustrating a second distributed speech recognition method in a telematics terminal according to a second embodiment of the present invention.

The present invention relates to an information providing system of a telematics system. In particular, the present invention relates to a telematics system using a distributed speech recognition system capable of simultaneously performing a remote speech recognition and a local speech recognition at a remote location and selecting a result that is quickly or accurately recognized by a user. An information providing system and method for a telematics system for receiving information provided by a service center of the system.

In general, a telematics system is a compound word of telecommunication and informatics, and combines a mobile communication method, a location tracking method, and the internet with a vehicle to detect a vehicle accident or theft of a vehicle and drive a vehicle. It is a system that guides the route and provides various other information to the vehicle driver.

That is, the vehicle telematics system is a system for providing information through a vehicle based on a GPS using a mobile communication and a satellite positioning system (GPS: Global Positioning System (GPS)) satellite. Accordingly, the vehicle telematics system uses the GPS, the wireless communication network, and the Internet network to provide vehicle drivers with traffic information, response to emergency situations, remote vehicle diagnosis, Internet use (eg, financial transactions, news provision, and e- It will provide various mobile communication services such as e-mail transmission and reception.

Due to the characteristics of the vehicle, the above services are often used while driving. However, when the driver directly manipulates a telematics terminal installed in the vehicle to use or operate the services while driving, an accident may occur as the driver's field of view moves away from the front.

This problem requires a new input system. As an alternative to this problem, a voice recognition system will be applied to the telematics terminal.

In general, in order to implement a speech recognition system, a speech recognition database for speech recognition should be established. In order to increase the recognition rate of the speech recognition system, the database should be registered with a lot of voice data information such as voice phoneme data, voice feature data, and rhyme data. In order to register a lot of information, the capacity of the voice recognition database must be increased.

However, telematics terminals are relatively small devices that are limited in their storage capacity. Therefore, since the capacity of the speech recognition database of the speech recognition system in the telematics terminal is limited, the registered information is also reduced.

In addition, in order to perform speech recognition, when the recognition target word is more than five hundred words, a large amount of computation and memory resources of 3-4 Mbytes are required. In addition, the floating point operation is difficult in the calculation for speech recognition, so that the calculation is performed by converting to fixed point.

In order to reduce the burden of speech recognition, distributed speech recognition processing technology has been in the spotlight. The distributed speech recognition means that a terminal receiving a voice detects and characterizes a voice feature of an input voice signal (hereinafter referred to as "voice feature data"), and sends the voice feature data to a remote voice recognition server. And the terminal receives the recognition result, thereby reducing the resource burden of the telematics terminal.

When the distributed speech recognition technology is described in more detail, speech recognition performance is significantly degraded due to compression transmission and transmission error of speech in a wireless communication environment. Distributed speech recognition is a technology that improves speech recognition performance by extracting features of speech signals, converting them into digital data, that is, speech feature data, and transmitting them.

As described above, the conventional distributed speech recognition system relies on communication with a remote speech recognition server based on a network through a mobile communication network, and thus cannot be used or takes a long time when the situation of the mobile communication network or network is poor. There was this.

Accordingly, an object of the present invention is to provide a service in a telematics system using a distributed speech recognition system that simultaneously performs remote distributed speech recognition and local speech recognition and selects a result that is quickly or accurately recognized by a user. An information providing system and method of a telematics system for receiving information are provided.

Another apparatus of the present invention for achieving the above object; An information providing system of a telematics system, comprising: detecting a voice feature of the voice from a spoken voice word to generate voice feature data, generating and transmitting the voice feature data, and responding to the transmission of the voice feature data A telematics terminal that receives information about a language and notifies a user, a distributed speech recognition server that receives the speech feature data and performs speech recognition using the speech feature data to transmit a speech recognition recognition word, and the recognition A database server that receives a word and finds and transmits information corresponding to the recognized word, and receives the voice feature data from the telematics terminal, transmits the received voice feature data to the distributed voice recognition server, and inputs the recognized word in response to the transmission of the voice feature data To the database server The web server is configured to transmit the information to the telematics terminal when information about the recognized word is received from the database server in response to the recognized word transmission.

The present invention for achieving the above other object; In the information providing system of the telematics system, the voice feature of the voice is detected from the spoken voice word to generate the voice feature data, and the voice feature data is generated to perform local voice recognition and to recognize the local recognition word as a recognition result. Generating and storing the voice feature data, receiving and storing a distributed recognizer in response to the transmission of the voice feature data, transmitting one recognizer of the local recognizer and the distributed recognizer, and transmitting the recognizer And a telematics terminal for receiving information on the recognized word and notifying the user in response thereto, and a distributed voice for receiving the voice feature data and performing voice recognition using the voice feature data to transmit a voice recognition distributed recognition word. A recognition server and the information corresponding to the received recognition word; A database server which finds and transmits the voice feature data from the telematics terminal and transmits the voice feature data to the distributed voice recognition server and transmits the voice feature data to the telematics terminal when the distributed recognizer inputs in response to the transmission of the voice feature data. The web server transmits the recognized word from the telematics terminal to the database server in response to the transmission, and transmits the recognized word to the telematics terminal when information about the recognized word is received from the database server.

The method of the present invention for achieving the above object; An information providing method of a telematics system comprising a telematics terminal, a web server, a distributed speech recognition server, and a database server, wherein the telematics terminal detects a voice feature of the voice from a spoken voice word and generates voice feature data. Transmitting the voice feature data to the distributed voice recognition server; transmitting the voice feature data from the telematics terminal to the distributed voice recognition server; and receiving the voice feature data and receiving the voice feature data. Performing a voice recognition using data to transmit a recognition word that is a result of speech recognition to the database server, the database server finding information corresponding to the recognition word, and transmitting the information to the web server, and the web server The telematics terminal with the information Transfer process and further characterized in that said telematics terminal, constituted by any process, for receiving information on the voice control in response to the transmission of the speech feature data, notifying the user.

The method of the present invention for achieving another object of the present invention; An information providing method of a telematics system comprising a telematics terminal, a web server, a distributed speech recognition server, and a database server, wherein the telematics terminal detects a voice feature of the voice from a spoken voice word and generates the voice feature data. Generating the voice feature data, performing local voice recognition, transmitting the voice feature data, and receiving, by the web server, the voice feature data and transmitting the voice feature data to the distributed voice recognition server; A distributed speech recognition server, receiving the speech feature data and performing speech recognition using the speech feature data to transmit a speech recognition distributed recognition word to the web server; Transmitting to the telematics terminal, the telematics terminal, the Transmitting, to the web server, any one of a distributed recognition word and a local recognition word that is a result of performing the local speech recognition recognition; transmitting, by the web server, the recognition word to the database server; Receiving, by the database server, and searching for and transmitting information corresponding to the recognized word; transmitting, by the web server, the information to a telematics terminal; and transmitting, by the telematics terminal, the recognized word. And receiving information on the recognized word in response to the notification.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that in the following description, only parts necessary for understanding the operation according to the present invention will be described, and descriptions of other parts will be omitted so as not to depart from the gist of the present invention.

The present invention simultaneously performs distributed speech recognition through a distributed speech recognition server at a remote location and speech recognition by the telematics terminal itself (hereinafter referred to as "local speech recognition") and performs the distributed speech recognition result (hereinafter referred to as "distribution"). "Recognition word") and the local voice recognition result (hereinafter referred to as "local recognition word") to determine whether the match is notified to the user, and if it is different, all the recognition results are output and can be selected by the user It is.

In addition, the present invention is to simultaneously perform distributed speech recognition through the remote speech recognition server and the telematics terminal itself speech recognition at the time of speech recognition in the telematics terminal and to inform the user of the result of the faster speech recognition in time.

In addition, according to the first embodiment of the information request, the present invention transmits the information on the speech recognized recognition word to a service center and receives the information on the recognition word to inform the user.

In addition, according to a second embodiment of the information request, the present invention performs voice transplantation when a request for information on the recognized word is performed and finds information corresponding to the recognized speech word from a database and transmits the information directly to the telematics terminal. .

In addition, the present invention, as described above, the telematics terminal receives the distributed identifiers recognized from the service center, and transmits the corresponding identifiers to the service center by determining whether they match the local identifiers and determining by the user. To receive.

1 is a diagram illustrating a configuration of a distributed speech recognition system of a telematics system according to the present invention.

The distributed speech recognition system of the present invention comprises a telematics terminal 100, a mobile communication system, and a service center 200 connected through an internet network or connected through a mobile communication system. In the following description, it should be noted that the method of transmitting and receiving data and the method of transmitting and receiving data through the mobile communication network and the Internet network will not be described or mentioned for convenience of description by the person skilled in the art.

The telematics terminal 100 detects a voice feature from the spoken voice word, generates voice feature data, and transmits the voice feature data to the service center 200. In addition, the telematics terminal 100 performs voice recognition by itself based on the characteristic data. When the recognition result is received from the service center 200, the telematics terminal 100 compares the self recognition result when the recognition result is received from the service center 200, and informs the user of one result if it is the same. In response, when the user selects one of them, the selected recognition result is set as the final recognizer.

The service center 200 receives the voice feature data through the internet network, performs voice recognition by the voice feature data, finds information on the recognized recognition word, and transmits the information to the telematics terminal 100, or The recognized recognition word is transmitted to the telematics terminal 100, and in response thereto, information on the received recognition word is found and transmitted to the telematics terminal 100.

Specifically, the service center 200 includes a web server 210, a distributed voice recognition server 220, and a database server (hereinafter referred to as “DB”) server 230.

The web server 210 interfaces data between the telematics terminal 100, the distributed speech recognition server 220, and the DB server 230. That is, when the message received from the telematics terminal 100 is a voice recognition request message including voice feature data, the message is transmitted to the distributed voice recognition server 220, and when the information request message requests specific information, the DB server 230. To send). The distributed speech recognition server 220 includes a large speech recognition DB, receives the speech recognition request message through the web server 210, detects speech feature data from the speech recognition request message, and the speech. After analyzing the feature data, the voice recognition is performed by referring to the voice recognition DB, and the voice recognition recognition word is transmitted to the web server 210. The web server 210 receives the recognized word and transmits it to the telematics terminal 100 according to the first request for information request and sends an information request message for the recognized word according to the second request for information request to the DB server ( 230). The information request according to the first embodiment of the information request transmits a recognition word received from the distributed speech recognition server 220 to the telematics terminal 100 and then responds to the recognition word received in response to the DB server 230. Send an information request message.

The DB server 230 receives the information request message from the web server 210 and finds information on the recognized word included in the information request message and transmits the information to the web server 210. Then, the web server 210 transmits the information about the recognized word to the telematics terminal 100.

2 is a view showing the configuration of a telematics terminal according to the present invention. Hereinafter, the configuration of the telematics terminal according to the present invention will be described.

The telematics terminal 100 includes the control unit 10, the storage unit 20, the input unit 30, the audio processor 40, the voice recognition module 50, and the distributed speech recognition (hereinafter referred to as “DSR”). It consists of a processing unit 60 and a wireless communication unit 70.

The control unit 10 controls the overall operation of the telematics terminal according to the present invention.

The storage unit 20 is an area for storing a control program for distributed speech recognition and information on a recognized recognition word in a telematics system, an area for temporarily storing data generated during execution of the control program, and a telematics. And a user data storage area for storing user data such as a function of a terminal or a music file.

The input unit 30 generates and outputs a plurality of input data for controlling a function of a telematics terminal by a user as an input device such as a key input device and / or a touch screen.

The display unit 35 displays various information such as a moving image, a still image, and the like under the control of the controller 10.

The audio processor 40 is connected to a speaker SP or a microphone MIC, and converts an analog voice word input through the microphone into voice data in a digital form under the control of the controller 10 and the voice recognition module. Or 50 and / or to the DSR processing unit 60.

The voice recognition module 50 has the small voice recognition DB under the control of the controller 10, receives the voice data, detects the voice feature of the voice, and generates voice feature data. When the voice feature data is generated, the voice recognition module 50 performs the voice recognition using the voice feature data and the information of the voice recognition DB and outputs a recognition word that is a recognition result to the controller 10.

The DSR processor 60 receives the voice feature data from the audio processor 40 under the control of the controller 10 and controls an overall operation for distributed speech recognition processing through the wireless communication unit 70.

The wireless communication unit 70 forms a wireless channel with a mobile communication system under the control of the control unit 10, transmits data input from the control unit 10 or the DSR processing unit 60, and transmits the received data to the control unit 10. ) Or to the DSR processing unit 60.

3 is a flowchart illustrating a distributed speech recognition method of the telematics system according to the first embodiment of the information request. Hereinafter, a first embodiment for requesting information will be described with reference to FIG. 3.

First, when the voice recognition input data is input from the input unit 30, the telematics terminal 100 receives a voice through the microphone MIC. When a voice is input through the microphone MIC, the controller 10 controls the audio processor 40 and the voice recognition module 50 to output voice feature data to the DSR processor 60. Then, the DSR processor 60 generates a voice recognition request message including the voice feature data, and transmits the voice recognition request message to the web server 210 in step 251.

The web server 210 receiving the voice recognition request message transmits the voice recognition request message to the DSR server 220 in step 253.

Then, the DSR server 220 detects voice feature data from the voice recognition request message and performs voice recognition using the voice feature data to generate a recognition word that is a voice recognition result. When the recognition word is generated, the DSR server 220 transmits a voice recognition result message including the recognition word to the web server 210 in step 255.

Then, the web server 210 transmits the voice recognition result message to the DB server 230 in step 257.

The DB server 230 receiving the voice recognition result message detects a recognized word from the voice recognition result message and finds information on the recognized word. If information about the recognized word is found, the DB server 230 transmits the information about the recognized word including the found information to the web server 210 in step 259.

Then, the web server 210 transmits to the telematics terminal 100 in step 261.

4 is a flowchart illustrating a distributed speech recognition method of a telematics system according to a second embodiment of the information request of the present invention. A second embodiment for requesting information will be described with reference to FIG. 4.

First, when the voice recognition input data is input from the input unit 30, the telematics terminal 100 receives a voice through the microphone MIC. When a voice is input through the microphone MIC, the controller 10 outputs voice feature data to the DSR processor 60 through the audio processor 40 and the voice recognition module 50. Then, the DSR processor 60 generates a voice recognition request message including the voice feature data. When the voice recognition request message is generated, the DSR processor 60 transmits the voice recognition request message to the web server 210 in step 321. The telematics terminal 100 performs local voice recognition through the voice recognition module 50 after transmitting the voice recognition request message.

The web server 210 transmits the received voice recognition request message to the DSR server 220 in step 323.

Upon receiving the voice recognition request message, the DSR server 220 detects voice feature data from the voice recognition request message and performs voice recognition. The DSR server 220 performs voice recognition, and transmits a voice recognition result message including the recognition word as a result to the web server 210 in step 325. In step 327, the web server 210 transmits the voice recognition result message to the corresponding telematics terminal 100. The telematics terminal 100 receives the voice recognition result message, compares a local voice recognition word that is a result of performing local voice recognition with a distributed recognition word included in the voice recognition result message, and includes any one of the two if the same. An information request message for the recognized word is generated and transmitted to the web server 210 in step 329. In step 331, the web server 210 transmits an information request message for the recognized word to the DB server 230. Then, the DB server 230 finds information about the recognized word of the information request message, generates an information message about the recognized word, and transmits the information message to the web server 210 in step 333. The web server 210 receives the information message on the recognized word and transmits it to the telematics terminal 100 in step 335. The telematics terminal 100 for the recognized word informs the user through a display device (not shown) such as a liquid crystal display or the like and / or a speaker of the audio processor 40.

In the second embodiment of the information request, the telematics terminal receives both a local recognition word that is a result of performing its own voice recognition and a distributed recognition word through the DSR server 220. Hereinafter, a method of processing the local identifier and the distributed identifier in the telematics terminal will be described with reference to FIGS. 5 and 6.

5 is a flowchart illustrating a first distributed speech recognition method in a telematics terminal according to the first embodiment of the speech recognition.

The controller 10 determines whether voice is input through the microphone MIC and the audio processor 40 in step 401. In step 401, the voice input to the audio processor 40 is converted into voice data and output to voice recognition module 50.

After step 401, the controller 10 controls the voice recognition module 50 to detect voice feature data from the voice data and outputs the detected voice feature data to the DSR processor 60 in step 403. When the voice feature data is output to the DSR processor 60, the controller 10 outputs a control signal requesting to transmit the voice feature data to the DSR processor 60 in step 405. Then, the DSR processor 60 generates a voice recognition request message including the received voice feature data and transmits the generated voice recognition request message to the web server 210 through the wireless communication unit 70 and the antenna ANT. The DSR processor 60 checks whether a voice recognition result message is received in response to the transmission of the voice recognition request message through the wireless communication unit 70, and distributes the voice recognition result message from the voice recognition result message when the voice recognition result message is received. The recognition word is detected and output to the controller 10.

After outputting the transmission request control signal of the voice feature data, the controller 10 controls the voice recognition module 50 to perform local voice recognition in step 407.

After step 407, the controller 10 checks whether a distributed recognition word is input from the DSR processor 60 in step 409. If the distributed recognition word is not input, the process proceeds to step 413 where local voice recognition is completed and it is determined whether the local recognition word is input from the voice recognition module 50. When the distributed recognition word is input in step 407, the controller proceeds to step 411 to store the distributed recognition word. When the local recognition word is input from the speech recognition module 50 in step 409, the local recognition word is stored in step 415. do.

After step 411 or step 415, the control unit 10 proceeds to step 417 to check whether both the local recognizer and the distributed recognizer are stored.

The process after step 409 is repeated until both the local recognition word and the distributed recognition word are stored. If both the local recognition word and the distributed recognition word are stored in step 417, the controller 10 proceeds to step 418 and checks whether the local recognition word and the distributed recognition word match. If the local recognition word and the distributed recognition word match, the controller 10 proceeds to step 423 and generates an information request message for any one of the two recognition words and transmits it to the web server 210 through the wireless communication unit 70. do.

On the other hand, if the local recognition word and the distributed recognition word do not match, the controller 10 proceeds to step 419 and outputs the two recognition results through the audio processing unit 40 and / or display unit 35 and asks the user to select. After the selection request, the controller 10 proceeds to step 421 and checks whether one of the two is selected. If one of the two is selected, the controller 10 proceeds to step 423 and generates an information request message for the selected recognition word. 70) to the web server 210.

6 is a flowchart illustrating a second distributed speech recognition method in a telematics terminal according to a second embodiment of speech recognition of the present invention.

Referring to FIG. 6, the control unit 10 performs the same operation as in step 401 to step 415 of FIG. 5 in step 501 to 515 of FIG. 5. However, in the second embodiment, as shown in step 418 of FIG. 5, the web server inputs a recognition word that is input first among local recognition words and distributed recognition words without determining whether the two recognition words match. Transmit to 210. In the second embodiment, a recognized word input later is discarded.

As described above, the present invention can receive various types of information of the service center of the telematics system by voice, so that words, phrases, clauses, or top-down menu selection schemes for information that the driver wants to receive while driving are used. Since there is no need to click the touch screen when searching for information, there is an advantage that can help the driver to drive safely.

In addition, the present invention has the advantage that the recognition rate can be improved by performing distributed recognition using a distributed recognition system using a large database and local recognition of the terminal itself.

Claims (9)

In the information providing system of the telematics system, Detects the voice feature of the voice from the spoken voice word to generate voice feature data, generates and transmits the voice feature data, and receives information about the voice word in response to the voice feature data transmission to the user. A telematics terminal for notifying, A distributed speech recognition server for receiving the speech feature data and performing speech recognition using the speech feature data to transmit speech recognition recognition words; A database server which receives the recognition word and finds and transmits information corresponding to the recognition word; Receives voice feature data from the telematics terminal and transmits the voice feature data to the distributed voice recognition server. When the recognition word is input in response to the transmission of the voice feature data, the voice feature data is transmitted to the database server and from the database server in response to the recognition word transmission. And a web server transmitting the information to the telematics terminal when the information on the recognized word is received, the information providing system of the telematics system using the distributed speech recognition system. In the information providing system of the telematics system, Detect the voice feature of the voice from the spoken voice word to generate the voice feature data, generate the voice feature data to perform local voice recognition, generate and store a local recognition word as a result of the recognition, and store the voice feature data Transmits and receives and stores a distributed recognition word in response to the transmission of the voice feature data, transmits one of the local recognition word and the distributed recognition word, and transmits the information on the recognition word in response to the transmission of the recognition word. And a telematics terminal for receiving a notification to the user, A distributed speech recognition server for receiving the speech feature data and performing speech recognition using the speech feature data to transmit a speech recognition distributed recognition word; A database server which receives the recognition word and finds and transmits information corresponding to the recognition word; Receives voice feature data from the telematics terminal and transmits the voice feature data to the distributed voice recognition server. When the distributed recognizer inputs the voice feature data in response to the voice feature data transmission, the voice feature data is transmitted to the telematics terminal. Information of a telematics system using a distributed speech recognition system, comprising: a web server transmitting the recognized word from a terminal to the database server; and transmitting the recognized word from the database server to the telematics terminal. Provide system. The method of claim 2, And the telematics terminal transmits the first recognized recognition word among the local recognition word and the distributed recognition word to the web server. The method of claim 2, And the telematics terminal determines whether the local recognition word and the distributed recognition word match and transmits any recognition word to the web server. The method of claim 4, wherein The telematics terminal informs the user of the local recognition word and the distributed recognition word if the local recognition word and the distributed recognition word do not match, and transmits a recognition word selected by the user to the web server. Information provision system of telematics system using. In the information providing method of a telematics system comprising a telematics terminal, a web server, a distributed speech recognition server and a database server, Detecting, by the telematics terminal, a voice feature of the voice from the spoken voice word, generating voice feature data, and transmitting the voice feature data to the web server; Receiving, by the web server, voice feature data from the telematics terminal and transmitting the voice feature data to the distributed voice recognition server; Receiving, by the distributed speech recognition server, the voice feature data and performing voice recognition using the voice feature data to transmit a recognition word that is a voice recognition result to the database server; Transmitting, by the database server, information corresponding to the recognized word to the web server; Transmitting, by the web server, the information to the telematics terminal; And receiving, by the telematics terminal, information on the voice word in response to the transmission of the voice feature data, and notifying the user of the information of the telematics system. In the information providing method of a telematics system comprising a telematics terminal, a web server, a distributed speech recognition server and a database server, The telematics terminal detecting the voice feature of the voice from the spoken voice word to generate the voice feature data, generating the voice feature data to perform local voice recognition, and then transmitting the voice feature data; Receiving, by the web server, the voice feature data and transmitting the received voice feature data to the distributed voice recognition server; Receiving, by the distributed speech recognition server, the speech feature data and performing speech recognition using the speech feature data to transmit the speech recognized distributed recognition word to the web server; Transmitting, by the web server, the distributed recognition word to a telematics terminal; Transmitting, by the telematics terminal, any one of the distributed recognition words and local recognition words that are the result of performing the local speech recognition recognition to the web server; Transmitting, by the web server, the recognized word to the database server; Receiving, by the database server, the recognized word and finding and transmitting information corresponding to the recognized word; Transmitting, by the web server, the information to a telematics terminal; And receiving, by the telematics terminal, information about the recognized words in response to the transmission of the recognized words, and notifying the user to the information of the telematics system using the distributed speech recognition system. The method of claim 7, wherein And the telematics terminal transmits the first recognized recognition word among the local recognition word and the distributed recognition word to the web server. The method of claim 7, wherein And the telematics terminal determines whether or not the local recognition word and the distributed recognition word match, and transmits any recognition word to the web server.

Images