KR20060096654A - Mobile service system using multi-modal platform and method thereof - Google Patents

Abstract

The present invention discloses a mobile service system and a method using a multi-modal platform that can be applied to a specific specific mobile community service to converge and provide voice and text information. The system includes a mobile communication terminal for accessing the Internet through a swap / web browser and transmitting voice data and a current Uniform Resource Locator (URL); A multi-modal platform for mapping a speech recognition grammar using a current URL transmitted from the mobile communication terminal and generating a target URL of a new page to be moved; An Automatic Speech Recognition (ASR) server for recognizing speech using speech data and speech recognition grammar transmitted from the multi-modal platform; A mobile community server that registers contents of a swap or multimodal community application by the multimodal platform according to a voice request of a subscriber and receives the result and delivers the result to the mobile communication terminal; And a web server which receives a voice content registration request according to the voice of the subscriber from the mobile community server and returns the result.

Multi-Modal, Cyworld, Community, Mobile, Voice Recognition

Description

Mobile service system using multi-modal platform and method thereof {Mobile Service System Using Multi-Modal Platform And Method Thereof}

The following drawings, which are attached in this specification, illustrate exemplary embodiments of the present invention, and together with the detailed description of the present invention, serve to further understand the technical spirit of the present invention. It should not be construed as limited to.

1 is a schematic configuration diagram of a mobile service system using a multi-modal platform according to the present invention.

2 is a block diagram showing the configuration of a mobile communication terminal.

3 is a block diagram showing the configuration of a server for voice recognition of FIG.

4 is a flowchart illustrating a mobile service method using a multi-modal platform according to the present invention.

5 is a flowchart illustrating a mobile service method according to another embodiment of the present invention.

<Description of Signs for Main Parts of Drawings>

10: mobile terminal 11: microphone

12: EVRC Encoder 13: Multimodal Module

14: swap / web browser 15: wireless module

20: exchange 30: multimodal platform

31: network connection unit 32: voice data conversion unit

33: speech recognition grammar mapping unit 34: global grammar mapping unit

35: target URL generator 36: database

40: Automatic Speech Recognition (ASR) Server

50: mobile community server 60: web server

The present invention relates to a wireless swap service using a multi-modal platform, and more particularly, a mobile service system using a multi-modal platform that can provide multi-modal technology to a specific mobile community service and provide voice and text information. It's about how.

With the spread of mobile communication terminals, the necessity of using information on the Internet is increasing in mobile communication terminals, and the performance of hardware, speed of communication network, screen size, input device, etc. of mobile communication terminals are higher than those of PCs.

Due to the significant difference in performance and constraints, there were many problems in using the Internet browser and contents for the conventional PC and the wired Internet network. However, this problem has been solved through new content formats and input methods (eg, navigation using numeric buttons), such as WAP, but there are limitations in showing many menus and links at the same time on a small screen of a mobile communication terminal. In order to move from the menu of the initially set web page to the web page with the data desired by the user, it is necessary to continuously input specific keys along the linked links to reach the final content. Can be. Therefore, there is a problem in that the time required for connection is increased because a special function key must be clicked in order to continuously select characters along the tree and input of characters such as URLs.

Accordingly, there has been an attempt to solve the inconvenience caused by the connection as described above through the mobile communication terminal with a built-in voice recognizer.

However, the mobile terminal capable of voice recognition should have a separate voice recognition module, and the mobile terminal capable of voice recognition can process only the same voice data as the voice command of the built-in memory. Since there is a need to consume the resources of the storage means and have a separate module, there is a problem in that the space inside the mobile communication terminal is consumed. Therefore, voice XML and voice recognition technology are used to provide services through commands or voice, without the limitation of narrow screen and limited key input of mobile communication terminal.

Meanwhile, one-person media and community services (eg, Nate's Cyworld, Naver's blog, etc.), which are activated by wire, are being applied to mobile. Therefore, the subscriber can view and post his own minihompy using the mobile swap, and can also query the minihompy of others. Such community service applied to mobile may be a WAP-based service and a mobile phone application.

Research on a service that can fuse voice and text information by applying a multi-modal technology to the mobile-based community service is in progress.

Accordingly, an object of the present invention is to solve the above problems of the prior art, an object of the present invention by applying a multi-modal technology to a specific mobile community service by providing a converged voice and text information, uncomfortable information through the keyboard The present invention provides a mobile service system and a method using a multi-modal platform that can easily post by voice and overcome an inconvenient input limit without posting.

Mobile service system using a multi-modal platform of the present invention according to the first aspect for achieving the above object, the mobile communication to access the Internet through a swap / web browser, and to transmit voice data and the current Uniform Resource Locator (URL) Terminal; A multi-modal platform for mapping a speech recognition grammar using a current URL transmitted from the mobile communication terminal and generating a target URL of a new page to be moved; An Automatic Speech Recognition (ASR) server for recognizing speech using speech data and speech recognition grammar transmitted from the multi-modal platform; A mobile community server that registers contents of a swap or multimodal community application by the multimodal platform according to a voice request of a subscriber and receives the result and delivers the result to the mobile communication terminal; And a web server which receives a voice content registration request according to the voice of the subscriber from the mobile community server and returns the result.

According to a second aspect of the present invention, there is provided a mobile service system using a multimodal platform of the present invention, a browser accessing the Internet through a wireless network, an encoder converting voice data input into a microphone, and a current connection from the browser. A mobile communication terminal having a multi-modal module for transmitting the URL information of the site, the voice data converted by the encoder, and the information of the mobile communication terminal to a multi-modal server; A target URL to determine a grammar required for voice recognition from the current URL information transmitted from the mobile communication terminal, transmit the voice recognition grammar to the ASR server together with the voice data, and access the target URL through a result recognized by the voice recognition server Generating a multi-modal platform for transmitting to the mobile communication terminal; An ASR server for recognizing speech using speech data and speech recognition grammar transmitted from the multi-modal platform, and transmitting the recognized result to the multi-modal platform; And a web server to which the mobile communication terminal accesses using a target URL transmitted from the multi-modal platform.

In addition, in the mobile service method using a multi-modal platform according to the present invention, A) the controller of the mobile communication terminal determines whether a swap / web browser by a specific key is executed, and when the swap / web browser is executed, Converting and storing a voice input into a microphone of the terminal in an encoder and transmitting the stored information together with a URL of an initially set page; B) generating, by the multi-modal platform, a speech recognition grammar required for speech recognition using the URL of the current page transmitted from the mobile communication terminal, and transmitting the generated speech recognition grammar to the ASR server along with the speech data; C) the voice recognition server, recognizing the transmitted voice using the transmitted voice data and the voice recognition grammar, and transmits the recognized result to the multi-modal platform; D) the mobile community server, after receiving the subscriber's recorded voice message storage request message from the multi-modal platform, forwarding the content registration request according to the request message to a web server; And E) transmitting, by the web server, a response message according to the content registration request to the mobile communication terminal through the mobile community server and the multi-modal platform.

Therefore, according to the present invention, by applying a multi-modal technology to a specific mobile community service by providing a converged voice and text information, the application can be conveniently utilized by using the movement of another person's mini homepage or voice command based on voice recognition. Enable to use

In the reference numerals to the components of the accompanying drawings, it should be noted that the same reference numerals have the same reference numerals as much as possible even if displayed on different drawings. In addition, many specific details are set forth in the following description and in the accompanying drawings, in order to provide a more thorough understanding of the present invention. It will be apparent to those skilled in the art that the present invention may be practiced without these specific details. In addition, detailed description of well-known functions and configurations that may unnecessarily obscure the subject matter of the present invention will be omitted.

The present invention is to apply a multi-modal technology to a mobile-based community application, if you apply the multi-modal technology can easily post to the voice without posting information uncomfortable through the keyboard can exceed the uncomfortable input limit In addition, it makes it possible to use the application conveniently by using other people's homepage or voice command based on voice recognition.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to FIGS. 1 and 5 as follows.

1 is a diagram showing the configuration of a mobile service system using a multi-modal platform according to the present invention.

As shown therein, the system is connected to the Internet through a swap / web browser 14, the mobile communication terminal 10 for transmitting voice data and the current Uniform Resource Locator (URL), the switch 20 and And a multimodal platform 30 for mapping a speech recognition grammar using a current URL transmitted from the mobile communication terminal 10 and generating a target URL of a new page to be moved, and a voice transmitted from the multimodal platform 30. ASR (Automatic Speech Recognition) server 40 for recognizing speech using data and speech recognition grammar, and the content of the swap or multi-modal community application will be described later by the multi-modal platform 30 according to a subscriber's voice request. The mobile community server 50 registers with the web server 60 and receives the result and transmits the result to the mobile communication terminal 10, and the subscriber from the mobile community server 50. And a web server 60 for receiving the voice content registration request according to the voice and returning the result.

2 is a block diagram showing the configuration of a mobile communication terminal. The mobile communication terminal 10 includes a call unit (not shown), a microphone 11, an EVRC encoder 12, a multi-modal module 13, and a swap / web. A browser 14 and a wireless module 15.

The mobile communication terminal 10 has a built-in multi-modal application, and provides the subscriber's mini homepage inquiry and voice / text posting, voice / text viewing function.

The encoder 12 compresses and converts the voice input through the microphone 11, and the encoder 12 is 8 Kbps EVRC or 13 Kbps QCELP, and different types of encoders may be used according to the type of mobile communication terminal. Preferably, the encoder 120 is an 8 Kbps EVRC encoder.

The multi-modal module 13 includes voice data compressed and converted by the encoder 12, URL information of the current page detected by the browser 14, and terminal information (e.g., type of browser, mobile communication phone). Number, etc.) is transmitted to the voice recognition system (not shown) through the wireless module 15.

3 is a block diagram showing the configuration of a server for voice recognition of FIG.

The server for speech recognition includes a multimodal platform 30 for generating a target URL to be newly moved according to a unique speech recognition grammar and a recognition result, voice data transmitted from the multimodal platform 30, and voice. It is composed of an Automatic Speech Recognition (ASR) server 40 for recognizing speech using a recognition grammar and transmitting the recognition result to the multi-modal platform 30.

The multimodal platform 30 includes a network connection unit 31, a voice data conversion unit 32, a speech recognition grammar mapping unit 33, a global grammar mapping unit 34, and a target URL generation unit 35. And a database 36.

The network connection unit 31 connects to a mobile communication terminal (not shown), transmits and receives data, and preferably connects using a TCP / IP protocol.

The voice data conversion unit 32 is connected to the network connection unit 31, and converts the voice data transmitted from the network connection unit 31 so that the voice recognition engine can process, preferably converted to PCM format, If the speech recognition engine can directly process the compression format (for example, EVRC format) transmitted from the mobile communication terminal, the conversion process can be omitted.

The speech recognition grammar mapping unit 33 detects a valid speech recognition grammar valid for the page from the URL transmitted from the network connection unit 31 and maps the generated speech recognition grammar to the current page for a predetermined voice command. In relation to foot

Map the conditions to be generated. That is, the voice recognition grammar specifies a predetermined voice command to be generated for a predetermined URL, and the list of voice commands input from the URL becomes the voice recognition grammar. For example, if a voice name of a company (for example, Nate, Naver, etc.) is inputted on a page for outputting one-person media and community service, the list of company names related to the service becomes voice recognition grammar, and the voice The recognition grammar and the input URL (eg Cyworld (http://cyworld.nate.com), blog (http://blog.naver.com), etc.) are mapped.

The global grammar mapping unit 34 may have a unique voice recognition grammar determined by the voice recognition mapping unit 33 and a voice command that is commonly valid regardless of the current page of the browser (eg, help, bookmark). Maps to a speech recognition grammar.

The target URL generation unit 35 is connected to the network connection unit 31 and the ASR server 40 as a target URL to move the URL designated by the command using a command according to the transmitted terminal information and the voice recognition result. It generates and transmits to the network connection unit 31. For example, if the voice recognition of 'voice posting' on 'cyworld' or 'selection of text and voice messages left by others' is performed, the URL of the corresponding subworld of cyworld site can be viewed. Is set to the target URL.

The ASR (Automatic Speech Recognition) server 40 is connected to the multi-modal platform 30, recognizes the speech using the speech data and the speech recognition grammar transmitted from the multi-modal platform 30, in the present invention Uses known speech recognition systems.

In addition, the ASR (Automatic Speech Recognition) server 40, the voice is divided into a simple voice command to read the currently displayed menu item and a short voice command crossing the menu tree of several levels, the voice of the simple voice command In this case, the user reads a menu item on the screen, and in case of a compound word menu item, various alternative labels should be considered. For example, a menu such as "Check for voice messages left by others" may allow users to speak short words, such as "Voices of other people." Is added to).

On the other hand, in the case of a short voice command, the voice pattern (Corpus) of the user who uses the application of the terminal is collected and the command is issued at once without going through a menu step. For example, for voices such as "selection of voice message list left by others", a specific phrase "list selection" is set as well as "voice message" for user convenience. In order to support the shortened voice command, the ASR server 40 should be capable of recognizing continuous words, and the grammar may include a grammatical structure of an Augmented Backus-Naur Form (ABNF) form or an equivalent form. Implemented to accommodate.

FIG. 4 is a flowchart illustrating a mobile service method using a multi-modal platform according to the present invention, illustrating a voice posting process of a subscriber.

In the following embodiment, the subscriber performs a voice command using the terminal's swap / web browser 14, whereby the URL of a page for outputting a single media and community service (e.g., Cyworld (http: // cyworld.nate.com), blogs (http://blog.naver.com), etc.), and check the URL by selecting 'your own voice posting' or 'list of text and voice messages left by others'. Explain the process of performing '.

First, the controller (not shown) of the mobile communication terminal 10 determines whether the swap / web browser 14 is executed by a specific key (S40), and when the swap / web browser 14 is not executed. When performing the general function of the mobile communication terminal (S41) and the swap / web browser 14 is executed, the encoder converts and stores the voice input into the terminal's microphone, and stores the stored information in the URL of the initially set page. (For example, Cyworld (http://cyworld.nate.com), blog (http://blog.naver.com)) and transmits to the multi-modal platform 30 (S42).

In step S42, when the swap / web browser 14 is executed, voice data input by the subscriber input to the mobile communication terminal 10, URL information of the current page, and the mobile communication terminal 10 are executed. ) Information is transmitted to the multi-modal platform (30).

In this case, it is determined whether an input is generated from the microphone 11 of the terminal, and if a voice input does not occur, general web surfing is performed. When a voice input occurs, the voice input from the microphone 11 is inputted to the encoder ( 12) and converts the multi-modal module 13 to the multi-modal module 13. When the converted voice data is transmitted, the multi-modal module 13 transfers the browser information and the URL of the current page from the currently running swap / web browser 14; Requesting information, and converting the converted voice data, the obtained URL information of the current page and mobile communication terminal information (for example, browser type, mobile communication phone number, etc.) through the wireless module 15 through the multimodal platform. To 30. However, since this is already known to those skilled in the art, further description will be omitted.

Thereafter, the multi-modal platform 30 generates a speech recognition grammar required for speech recognition using the URL of the current page transmitted from the mobile communication terminal 10, and generates the generated speech recognition grammar along with the voice data in the ASR server. Transfer to 40 (S43). At this time, the subscriber selects the recording menu for voice posting on a specific page of the URL and then records the voice.

In addition, in step S43, the multi-modal platform 30 is connected to the mobile communication terminal 10 requesting voice recognition and processes the voice data transmitted from the mobile communication terminal 10 in the voice recognition engine. Receives the current URL and the terminal information from the terminal at the same time as the PCM conversion, and maps the unique speech recognition grammar corresponding to each URL from the received current URL information, and the mapped unique speech recognition grammar You will combine valid global speech recognition grammars on the page.

Thereafter, the voice recognition server 40 recognizes the transmitted voice using the transmitted voice data and the voice recognition grammar, and transmits the recognized result to the multi-modal platform 30 (S44). At this time, the speech recognition method may use a known speech recognition method, preferably using a speaker-independent speech recognition method.

Thereafter, the mobile community server 50 receives a voice recognition result from the multi-modal platform 30, that is, a voice message of the subscriber (voice of the subscriber recorded for posting), and then stores the message (S45). The content registration request according to the request message is transmitted to the web server 60 (S46).

Thereafter, the web server 60 transmits a response message according to the content registration request to the mobile communication terminal 10 through the mobile community server 50 and the multi-modal platform 30 (S47). Therefore, through the above process, the subscriber recognizes that his voice message has been posted.

5 is an operation flowchart for explaining a mobile service method according to another embodiment of the present invention, and for explaining a voice posting process of the counterpart.

First, the control unit (not shown) of the mobile communication terminal 10 executes the swap / web browser 14 with a specific key, and then the URL of an initially set page (for example, Cyworld (http: //)). cyworld.nate.com), and a blog (http://blog.naver.com), transmit a voice message listening request message recorded by the counterpart to the multimodal platform 30 (S51).

Thereafter, the multi-modal platform 30 receives the voice message listening request message and transmits it to the mobile community server 50, thereby requesting a voice message patch (S52). Subsequently, the mobile community server 50 generates and transmits a target URL including a voice message recorded by the counterpart in response to the request, and simultaneously converts the voice of the counterpart converted by the voice data converter 32. For example, the EVRC format is transmitted to the mobile communication terminal 10 (S53).

Thereafter, the mobile communication terminal 10 displays the received URL page through the display device and the speaker and outputs the recorded voice of the other party to the subscriber (S54). Of course, the recorded voice of the other party is listed, and the subscriber may choose to cheat among them.

Therefore, the subscriber can watch voice and text messages recorded by the other party stored in his or her own minihomp without restriction of construction.

Although the present invention has been described as a specific preferred embodiment, the present invention is not limited to the above-described embodiments, and the present invention is not limited to the above-described embodiments without departing from the gist of the present invention as claimed in the claims. Anyone with a variety of variations will be possible.

Therefore, according to the present invention, the community service through the mobile is improved in quality.

In addition, the present invention simplifies the inconvenience of the input of the mobile device through the voice, and by providing a voice and a display at the same time, the degree of information delivery to the subscriber is greatly improved.

Claims (22)

A mobile communication terminal which accesses the Internet through a swap / web browser and transmits voice data and a current Uniform Resource Locator (URL); A multi-modal platform for mapping a speech recognition grammar using a current URL transmitted from the mobile communication terminal and generating a target URL of a new page to be moved; An Automatic Speech Recognition (ASR) server for recognizing speech using speech data and speech recognition grammar transmitted from the multi-modal platform; A mobile community server that registers contents of a swap or multimodal community application by the multimodal platform according to a voice request of a subscriber and receives the result and delivers the result to the mobile communication terminal; And And a web server for receiving a voice content registration request according to the voice of the subscriber from the mobile community server and returning the result. The method of claim 1, wherein the mobile communication terminal, Built-in multi-modal application, in order to provide the subscriber's homepage inquiry and voice / text posting, voice / text viewing function, An encoder which compresses and converts the voice input through the microphone and outputs the compressed voice; A multi-modal module configured to output voice data compressed and converted from the encoder, URL information of a current page detected from a swap / web browser, and terminal information; And A mobile service system using a multi-modal platform, characterized in that it comprises a wireless module for receiving the voice data and the respective information from the multi-modal and wirelessly transmitted to a voice recognition system. The method of claim 2, wherein the encoder, 8 Kbps EVRC or 13Kbps QCELP, a mobile service system using a multi-modal platform, characterized in that using different types according to the type of mobile communication terminal. The method of claim 2, wherein the information of the terminal, Mobile service system using a multi-modal platform, characterized in that any one of the type of swap / web browser, mobile phone number. The method of claim 1, wherein the mobile communication terminal and the muldy modal platform, A mobile service system using a multi-modal platform, characterized in that the voice of the subscriber and its voice recognition results are transmitted and received to each other via a TCP connection. The method of claim 1, wherein the automatic speech recognition (ASR) server, A mobile service system using a multi-modal platform, characterized in that the voice collected from the subscriber is distinguished by a simple voice command for reading the currently displayed menu item and a short voice command crossing the menu tree of several levels. The method of claim 1 or 6, wherein the Automatic Speech Recognition (ASR) server, When the voice input from the subscriber is a simple voice command, the user reads a menu item on the screen, and multi-modal is characterized in that various alternative labels (Alias) are added to Grammar for the user's convenience. Mobile service system using the platform. The method of claim 1 or 6, wherein the Automatic Speech Recognition (ASR) server, When the voice input from the subscriber is a short voice command, the continuous word recognition is implemented, and the Augmented Backus-Naur Form (ABNF) format is added to Grammar, wherein the mobile service system using a multi-modal platform. Wap / web browser accessing the Internet through a wireless network, encoder converting voice data input into the microphone, URL information of the site currently being accessed from the swap / web browser, voice data converted by the encoder and information of the terminal are transmitted. A mobile communication terminal having a multi-modal module; Determining a grammar required for speech recognition from the current URL information transmitted from the mobile communication terminal, and transmits a speech recognition grammar with the voice data, and generates a target URL to be accessed through the result recognized by the speech recognition server Multi-modal platform for transmitting to the mobile communication terminal; An ASR server for recognizing speech using speech data and speech recognition grammar transmitted from the multi-modal platform, and transmitting the recognized result to the multi-modal platform; A mobile community server that registers contents of a swap or multimodal community application by the multimodal platform according to a voice request of a subscriber and receives the result and delivers the result to the mobile communication terminal; And And a web server for receiving a voice content registration request according to the voice of the subscriber from the mobile community server and returning the result. The method of claim 9, wherein the encoder, 8 Kbps EVRC or 13Kbps QCELP, a mobile service system using a multi-modal platform, characterized in that using different types according to the type of mobile communication terminal. The method of claim 9, wherein the information of the terminal, Mobile service system using a multi-modal platform, characterized in that any one of the type of swap / web browser, mobile phone number. 10. The method of claim 9, wherein the mobile communication terminal and the muldy modal platform, A mobile service system using a multi-modal platform, characterized in that the voice of the subscriber and its voice recognition results are transmitted and received to each other via a TCP connection. The method of claim 9, wherein the automatic speech recognition (ASR) server, A mobile service system using a multi-modal platform, characterized in that the voice collected from the subscriber is classified into a simple voice command that reads a currently displayed menu item and a short voice command that traverses a menu tree of several levels. The method of claim 9 or 13, wherein the ASR (Automatic Speech Recognition) server, When the voice input from the subscriber is a simple voice command, the user reads a menu item on the screen, and multi-modal is characterized in that various alternative labels (Alias) are added to Grammar for the user's convenience. Mobile service system using the platform. The method of claim 9 or 13, wherein the ASR (Automatic Speech Recognition) server, When the voice input from the subscriber is a short voice command, the continuous word recognition is implemented, and the Augmented Backus-Naur Form (ABNF) format is added to Grammar, wherein the mobile service system using a multi-modal platform. A) the controller of the mobile communication terminal determines whether a swap / web browser is executed by a specific key, and when the swap / web browser is executed, the controller converts and stores the voice input into the terminal microphone in the encoder and stores the Transmitting the information along with the URL of the initially set page; B) generating, by the multi-modal platform, a speech recognition grammar required for speech recognition using the URL of the current page transmitted from the mobile communication terminal, and transmitting the generated speech recognition grammar to the ASR server along with the speech data; C) the voice recognition server, recognizing the transmitted voice using the transmitted voice data and the voice recognition grammar, and transmits the recognized result to the multi-modal platform; D) the mobile community server, after receiving the subscriber's recorded voice message storage request message from the multi-modal platform, forwarding the content registration request according to the request message to a web server; And E) the web server, the mobile service method using a multi-modal platform, characterized in that for transmitting the response message according to the content registration request to the mobile communication terminal through the mobile community server and the multi-modal platform. . The method of claim 16, wherein the voice of step (A) is A mobile service method using a multi-modal platform characterized by being divided into simple voice commands and short voice commands that cross a multi-level menu tree. The method of claim 16, wherein step (A) is A-1) multi-modal platform comprising the step of transmitting the voice data by the subscriber input to the mobile communication terminal, the URL information of the current page, the information of the mobile communication terminal to the multi-modal platform; Mobile service method used. The method of claim 18, wherein step (A-1), It is determined whether the input is generated from the microphone of the terminal, and if the voice input is not generated, general web surfing is performed. When the voice input occurs, the voice input from the microphone is converted by the encoder and transmitted to the multi-modal module. When the converted voice data is transmitted, the multi-modal module requests browser information and URL information of the current page from a currently running swap / web browser, and converts the converted voice data and the URL information of the obtained current page and the mobile communication terminal. Mobile service method using a multi-modal platform, characterized in that for transmitting information to the multi-modal platform through a wireless module. The method of claim 16, wherein step (A) is If the swap / web browser is not executed, the mobile service method using a multi-modal platform, characterized in that it further comprises the step of performing a general function of the mobile communication terminal. The method of claim 16, wherein step (B) is Subscriber, the mobile service method using a multi-modal platform, characterized in that the voice recording after selecting a recording menu for the voice posting on a specific page of the URL. The method of claim 16, wherein in step (B), The multi-modal platform is connected to the mobile communication terminal requesting voice recognition and receives the current URL and terminal information from the terminal while converting the PCM to perform voice recognition processing of the voice data transmitted from the mobile communication terminal. And mapping from the received current URL information to a unique speech recognition grammar corresponding to each URL, and combining the mapped unique speech recognition grammar with a global speech recognition grammar valid on any page. Mobile service method.

Images