KR100820217B1 - System and method for offering voice portal service using telephone - Google Patents

Abstract

The present invention relates to a system for providing a voice portal service using a telephone and a method thereof. When a user makes a call to a central control unit, a user accesses a specific web site and outputs a voice information signal by recognizing a DTMF signal transmitted from the user and a voice. The user accesses the Internet network through this voice information signal, retrieves the web document data of a predetermined web site, parses and analyzes it, and converts it into a voice format. The insufficient portion is corrected based on a database, and then the effective data is converted into a voice language for processing voice data and transmitted to the user as voice.

Voice Portal Service, Speech Recognition, Speech Formats, Web Parsing

Description

System and method for providing voice portal service using telephone {SYSTEM AND METHOD FOR OFFERING VOICE PORTAL SERVICE USING TELEPHONE}

1 is a block diagram of a system for providing a voice portal service using a telephone according to an embodiment of the present invention.

FIG. 2 illustrates a detailed configuration of the central control server of FIG. 2.

3 is a flowchart illustrating a method for providing a voice portal service using a telephone according to an embodiment of the present invention.

The present invention relates to a system for providing a voice portal service using a telephone and a method thereof, and more particularly, to a system for providing a voice portal service using a telephone for providing a voice portal service after a user accesses a web site using a regular telephone. It is about.

Computer Telephony Integration (CTI) is a technology for sharing information resources (data, fax, voice information, etc.) between different devices by connecting a phone and a computer, and simply sharing information resources between the phone and the computer. In addition, it provides the registered information by controlling the connected devices and forming a network with the existing information.

Such CTI technology is an information delivery means rather than an information search means. However, since the CTI technology does not utilize an existing information infrastructure, there is a limitation in using it as a shared medium such as a portal community.

The disadvantaged information group, such as the visually impaired, presbyopia, patients with disabilities, and users who have difficulty using the computer, have a very difficult process of accessing and acquiring information from various web sites on the Internet. There is a problem that is significantly behind in social competitiveness compared to these advantageous users.

Recently, the ARS system has been modified to deliver news articles, books, etc. by voice to the underprivileged through the telephone media. It is not enough to digest information that is distributed to a wide range of people offline.

Therefore, in view of the costs and social budgets that socially disadvantaged people have, as the underprivileged fall behind in social competitiveness, they can use all the information distributed on the Internet so that the underprivileged can regain their social adaptation and competitiveness. The need for information acquisition and accessibility should be assured.

The technical problem to be achieved by the present invention is to provide a voice portal service using a telephone that allows a user to access a specific web site on the Internet using a regular telephone or a mobile phone, and then retrieve and acquire information by voice through a convenient user interface. It is to provide a system and a method thereof.

The system for providing a voice portal service using a telephone according to the first aspect of the present invention, when a user makes a phone call and connects, accesses a predetermined website after attempting to connect to the Internet through user authentication through a user voice or a telephone button. A central control server for interpreting HTTP (HyperText Transfer Protocol) information and converting the information into a computer telephony integration (CTI) environment, thereby providing a voice portal service to enable the user to utilize all the Internet through a telephone network; And a database associated with the central control server to manage user subscription information, user setting information, user interface information, web access and analysis related data, and voice output data.

In the above, the central control server,

A data input unit which outputs a data input signal by recognizing a DTMF (Dual Tone Multi Frequency) signal or voice transmitted from the user through an automatic telephone recognition (ASR); A user manager for reading and inquiring user authentication and user setting information from the database through a data input signal transmitted from the data input unit; A voice synthesizer for synthesizing and outputting the DTMF signal, user voice, or text information as voice in accordance with voice synthesis information set for each user according to a query result of the user manager; Analyzing input information of the user through the data input unit, accessing a predetermined web site, and importing web document data that can be collected by HTTP information, parsing and analyzing the converted document into appropriate text information. A web analysis unit; And aligning text information transmitted from the web parsing unit to a telephone interface, and then writing the text information according to the IVR (Interactive Voice Response) and VoiceXML (Extensible Markup Language) standards for processing voice data of a voice and outputting the voice to a user's phone. It includes an output unit.

Wherein the central control server in response to the request of the web analysis unit optimizes the insufficient web elements of the web document data into text information, and further a web correction unit for transferring the relocation according to the text information to the web analysis unit Include.

The data input unit may include: a DTMF character combination module used for accessing another web site from the web site and inputting text on the web document; And a DTMF browsing module that performs a browsing operation such as searching, entering, transmitting, and user setting in a web document of the web site.

The DTMF character combination module includes a DTMF character table code table corresponding to a corresponding character for each of the telephone buttons, and the voice synthesis of the characters completed according to a telephone button input by a user through the DTMF character table code table. It is desirable to read aloud through wealth.

The DTMF browsing module preferably has a user interface for each phone button such as object / text movement, link object entry, object / character reading, web site termination, menu execution, shortcut key setting and execution in a web document.

The data input unit preferably includes a voice command list to enable the user to use the Internet such as movement, entry, and execution of a cursor in the web document.

The voice synthesizer may be individually set to speed, intonation, and volume according to taste.

The web analysis unit may include a command line analysis unit configured to analyze user input information through the data input unit and establish a suitable web information access system; A web access unit for accessing a web site according to the instruction of the command line analyzing unit; A data receiving unit storing web document data collectable by HTTP information from a web site accessed from the web accessing unit; A data inspecting unit for validating the web document data of the data receiving unit and processing normal / error according to the validity checking result; A data alignment unit for segmenting the web document data normally processed by the data inspection unit into individual units and rearranging the web document data according to one-dimensional information; A data filter unit for distinguishing data convertible into text information from the web document data rearranged by the data sorting unit, and requesting the web correction unit to inquire data that is not interpreted as the text information; A data inquiry unit for converting web document data transmitted from the data filter unit into text information and requesting an analysis of a web image entity; The web image object delivered from the data retrieval unit combines organic information such as internal information such as picture file name, tool tip information, caption information, location of web image, surrounding information, and lower page information connected to the web image. Web image analysis unit to interpret the text information; And a data transmission unit for assembling the web document data textified by the data retrieval unit and the web image analysis unit, and transferring the text information aligned to one-dimensional information to the data output unit.

The data output unit is interlocked with the speech synthesis unit to provide a user interface system, to accommodate user input information of the data input unit, and to be designed as a menu type IVR structure that can accommodate a large amount of command systems. Do.

On the other hand, the voice portal service providing method using a telephone according to the second aspect of the present invention, when the user makes a call to the central control server is connected to a predetermined web site to send and receive voice related voice portal service to the user through the telephone network A method, comprising: a) the central control server recognizing a DTMF (Dual Tone Multi Frequency) signal and a speech (Automatic Speech Recognition (ASR)) and outputting user input information; b) performing user authentication through the user input information transmitted in step a), accessing a web site by accessing an internet network, and then retrieving and parsing all web document data in the web site. Converting to a speech format after Parsing and analysis; c) searching whether the web document data parsed and analyzed in step b) is valid data for the user; And d) converting the valid data transmitted through steps b) and c) into a voice language for telephone voice data processing and transmitting the converted valid data to the user.

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.

First, a voice portal service is a service in which a voice recognition engine recognizes a voice for a user command and delivers Internet information required by the user to the user through voice.

The core technologies of voice portal service are Automatic Speech Recognition (ASR) technology, TextTo Speech (TTS) technology, and Language XML (VoiceXML) that converts HTML into voice information.

Next, a voice portal service providing system using a telephone according to an embodiment of the present invention will be described in detail with reference to FIGS. 1 and 2.

1 is a block diagram of a voice portal service providing system using a telephone according to an embodiment of the present invention, and FIG. 2 is a detailed block diagram of the central control server of FIG.

As shown in FIG. 1, in a voice portal service providing system using a telephone according to an embodiment of the present invention, when a user makes a phone call and is connected, the user attempts to connect to the Internet through user authentication through a user voice or a telephone button. Central control server 100 that provides voice portal service for users to utilize all internet through telephone network by connecting to web site, interpreting HTTP (HyperText Transfer Protocol) information and converting into computer telephony integration (CTI) environment And a database 200 associated with the central control server 100 to store and manage user subscription information, user setting information, user interface information, web access and interpretation related data, and voice output data. Does not.

First, as shown in FIG. 2, the central control server 100 may include a data input unit 110, a user manager 120, a voice synthesizer 130, a web analyzer 140, a web calibrator 150, and data. Including but not limited to the output unit 160.

The data input unit 110 is connected to the user through a specific telephone number (for example, 02) 736-6000, and then enters the user ID (ID) number and password together with the voice announcement. Do this.

In this case, the user manager 120 manages subscription accounts such as a user ID, a password, a social security number, a phone number, a favorite list, various browsing setting information, and user setting information such as voice synthesis information for each user.

The data input unit 110 and the user management unit 120 are automatically connected to the web site set as the start page according to the user's setting information when the user authentication is completed, and the web by the voice synthesis unit 130 when the web site is connected to the web site. Since the contents of the site are output on the telephone, the data input unit 110 performs mutual input / output using the telephone button DTMF and a user voice.

In this process, the data input unit 110 exchanges information with the data output unit 160 organically.

The data input unit 110 may be divided into a DTMF character combination module and a DTMF browsing module when a user inputs user input information by using a phone button. The DTMF character combination module may be used to access another web site on a web document. It is used when entering URLs or when entering text in text boxes in a web document, and the DTMF Browsing Module is used for searching, entering, sending, and user setting in web documents.

For example, the DTMF character combination module is automatically executed by a telephone button 5 in a text box in a web site, and reads a character that is completed in a phoneme unit in conjunction with the voice synthesizer 130 when a character is input with the button. give. The DTMF character combination module has a table corresponding to the corresponding character for each telephone button called DTMF character table code table itself.

Example 1) When the user enters "WWW", the corresponding telephone button presses "9 # 9 # 9 #".

The DTMF Browsing Module has the user interface shown in Example 2 below.

Example 2) In the phone button,

Key 8: Move focus down one object in the web document and read it.

Key 2: Move focus to one object in the web document and read it.

Key 5: Enter from link object (clickable) within web document.

Key 4: Move focus to the left of one character within the web document object and read it.

Key 6: Move the focus to the right of one character within the web document object and read it.

Key 3: Move focus below the column object in the web document and read it.

Key 9: Move focus over the column object in the web document and read it.

Key 1: Reread the object already pointed to within the web document.

Key 7: Find only the link object in the web document and move down and read.

Key 0: Closes the website.

* Key: Launches a context menu to set various browsing settings.

#Table key: Used to set and execute various shortcuts.

On the other hand, the data input unit 110 is operated in synchronization with the DTMF browsing module so that when the user input information is input using the user voice (ASR), the data input unit 110 is limited to browsing except for text input.

In order to input user input information using a user's voice, a user must be familiar with a voice command list. The voice command list is stored in the ASR DB of the database 200.

The process of synchronizing user voice and DTMF browsing module in a web document is shown in Example 3 below.

Example 3)

1. If you say "down" in your voice, the DTMF Browsing Module is synchronized with the # 2 key and moves the focus down one object in the web document and reads it.

2. If you say "enter" or "execute" in your voice, DTMF key 5 is activated and enters the link object within the web document.

3. When the user speaks "Real Estate", if there is a link in the web page called "Real Estate", the focus moves to that location.

The speech synthesizer 130 reads all the processes in the speech synthesis (TTS) at the time of DTMF input and speech output. The speed, intonation, and volume of the speech synthesis (TTS) can be individually set according to a user's taste. have.

The speech synthesis unit 130 automatically converts the speech information to the data output unit 160 by simply receiving the appropriately processed text information.

The web analysis unit 140 analyzes user input information, accesses a predetermined web site, and retrieves web document data that can be collected by HTTP information, parses and analyzes the data into appropriate text information. Output

The web analysis unit 140 is a command line analysis unit for analyzing the user input information through the data input unit 110 to establish a suitable web information access system, accessing a website according to the instructions of the command line analysis unit through the PSTN network It is necessary to check whether the web source arrives safely through the data access section and whether the web source can be accessed and processed through the data reception section. After performing the data validation, the data inspection unit that handles normal / error based on the validation result, and the data inspection unit divides the normally processed web source into individual units and rearranges the data according to the one-dimensional information from the two-dimensional information. Can be converted into text information among the rearranged web elements. Data filter unit which requests data to web correction unit 150 to inquire data that are not interpreted as text information and converts object property into meaningful text information among web elements delivered from data filter unit and converts web image object into web. Data inquiry unit for requesting analysis to the image analysis unit 140, internal information such as the image file name, tooltip information, caption information of the web image object delivered from the data inquiry unit, the position of the web image, the correlation with the surrounding information, The web image analysis unit which combines organic information such as sub-page information connected with the web image and interprets it as text information, and combines web document data that is textualized by the data retrieval unit and the web image analysis unit, And a data transmission unit that transmits the text information to the data output unit after inspecting the text information aligned with the one-dimensional information. The.

In the above, the web image interpreter is capable of tracking and textifying the surrounding point where the web image is dependently related so that the web image existing on the web page can be interpreted as text (word or sentence) representing the image. Find clues.

The web image analysis unit converts the web image into a complete alternative text by correcting the expression based on the clue extracted by the tracking.

The web image interpreting step of the web image analyzing unit may be divided into seven steps as follows.

The first step is to receive and confirm the web image, and collect web sources of neighboring points where the web image is dependently associated. The second step is to determine the web image file name. The web image is extracted from the web source and stored in the buffer.

The third step is to determine the location of the web image and extract the location information from the web source and store it in a buffer. The fourth step is to identify the web source before and after the web image. The web image is identified at the front and rear portions of the web image at a predetermined level from the web source and stored in a buffer.

The fifth step is to identify the web image link page. The web image is attempted to access the web image based on the link information provided from the web source, and the document title and the document content are analyzed and stored in the buffer.

The sixth step is to identify the web image leading path and access the previous web page based on the path to reach the web image, analyze the document title, the document content, and store it in the buffer.

The seventh step is a web image registration text selection process, and among the data stored in the buffer, the most suitable text information that can represent the web image is selected.

Meanwhile, the web correction unit 150 optimizes the incomplete web elements by receiving the inquiry from the web analysis unit 140. The web correction unit 150 processes the HTML protocol used to implement the object properties and functions among the web elements into appropriate text information. The corresponding relocation information is retransmitted to the web analysis unit 140.

In this case, the relocation information of the web correction unit 150 plays an essential role in properly configuring the user interface menu system in the data output unit 160.

The data output unit 160 firstly sorts the text data suitable for the telephone voice format transmitted from the web analyzer 140 in a menu structure according to the telephone interface, and prepares and operates in accordance with the IVR and VXML standards.

At this time, the data output unit 160 is interoperable with the speech synthesis unit 130 to have a sophisticated user interface system, and to properly accommodate the command system of the data input unit 110.

In the data output unit 160, the IVR structure is typically a link method, but the present invention adopts a menu method, where the menu method provides a most reasonable structure that can effectively accommodate a large number of command systems.

In the present invention, the web site is an HTTP-based web site and includes not only the HTML standard but also all of "media player", "JavaScript / VBScript" and "flash" (for example, all web sites,,,).

Referring to Figure 3 will be described in detail the operation of the voice portal service providing system using a telephone according to an embodiment of the present invention.

2 is a flowchart illustrating a method for providing a voice portal service using a telephone according to an embodiment of the present invention.

As shown in FIG. 2, in the method of providing a voice portal service using a telephone according to an embodiment of the present invention, when a user makes a call to the central control server 100 using a general telephone or a mobile phone, not a computer (S1). The central control server 10 requests an ID and a password to the accessed user, and performs user authentication when the user inputs his or her ID and password. (S2)

When the user authentication is normally completed, the user speaks the address input window by voice or presses a telephone button (for example, *, 2, 5) in order to input a URL through the data input unit 110. The server 100 accesses the corresponding website according to the URL request of the user (S3), and loads all web sources belonging to the website, analyzes and corrects the object, and performs a conversion process into meaningful text.

That is, the data input unit 110 recognizes a DTMF signal or a user speech (Automatic Speech Recognition, ASR) and outputs a user input information signal through digital conversion.

The web analysis unit 140 accesses a real Internet network (PSDN) through a user input information signal, accesses a specific website, and loads all web document data of the website to parse and analyze the appropriate voice. Convert to format (TTS) (S5, S6).

At this time, the web analysis unit 140 searches whether the web document data is valid data for the user, and corrects the insufficient part of the data based on the web correction unit 150 (S7, S8).

Since the data output unit 160 processes the web document data into meaningful texts through the web analysis unit 140, the data output unit 160 converts the web document data into an appropriate voice format in conjunction with the voice synthesis unit 130 and outputs the voice to the user's phone. (S9, S10)

The drawings and detailed description of the invention are merely exemplary of the invention, which are used for the purpose of illustrating the invention only and are not intended to limit the scope of the invention as defined in the appended claims or claims. Therefore, those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

As described above, the system and method for providing a voice portal service using a telephone according to the present invention make it possible to easily access a website using information by using a telephone to an ordinary person, especially an information marginalized layer, and to call a company operating a website. It is effective to meet the call center system.

In addition, the voice portal service providing system and method using the telephone according to the present invention can be interlocked with the home network system in the future as the telephone utilization and the Internet utilization is expanded, and the alternative means for using information is expanded to rotate information distribution throughout society This has the effect of making it happen quickly.

Claims (14)

delete When a user makes a phone call and connects to a website after authenticating the user through a user voice or telephone button, attempting to connect to the Internet, and interpreting HTTP (HyperText Transfer Protocol) information, the computer telephony integration (CTI) environment A central control server for providing a voice portal service so that the user can utilize all the Internet through a telephone network; And a database for managing user subscription information, user setting information, user interface information, web access and analysis related data, and voice output data in association with the central control server. The central control server, A data input unit which outputs a data input signal by recognizing a DTMF (Dual Tone Multi Frequency) signal or voice transmitted from the user through an automatic telephone recognition (ASR); A user manager for reading and inquiring user authentication and user setting information from the database through a data input signal transmitted from the data input unit; A voice synthesizer for synthesizing and outputting the DTMF signal, user voice, or text information as voice in accordance with voice synthesis information set for each user according to a query result of the user manager; Analyzing input information of the user through the data input unit and accessing a predetermined website (WEB SITE), and importing web document data that can be collected by HTTP information, parsing and analyzing and converting it into appropriate text information. A web analysis unit; A web correction unit for optimizing insufficient web elements of the web document data into text information according to a request of the web analysis unit, and transferring the relocation corresponding to the text information to the web analysis unit; And After the text information transmitted from the web analysis unit is aligned to the telephone interface, data is outputted to the user's telephone by writing in accordance with the Interactive Voice Response (IVR) and VoiceXML (Extensible Markup Language) standards for processing the telephone voice data. Voice portal service providing system using a telephone, characterized in that configured to include. delete The method of claim 2, The data input unit, A DTMF character combination module used for accessing the web site from another web site and inputting text on the web document; And a DTMF browsing module that performs a browsing operation such as searching, entering, transmitting, and user setting in a web document of the website. The method of claim 4, wherein The DTMF character combination module includes a DTMF character table code table corresponding to a corresponding character for each of the telephone buttons, and the voice synthesis of the characters completed according to a telephone button input by a user through the DTMF character table code table. Voice portal service providing system using a telephone, characterized in that the voice is read through the unit. The method of claim 4, wherein The DTMF browsing module is a text / object in the web document for each telephone button. Move, Enter Link Object, Read Object / Character, Exit Website, Launch Menu, Set Shortcut Key And a user interface such as execution. Voice portal service providing system. The method of claim 2, The data input unit is a voice portal service providing system using a telephone, characterized in that it has a voice command list to enable the use of the Internet, such as movement, entry, execution of the cursor in the web document to the user voice. The method of claim 2, The voice synthesizer is a voice portal service providing system using a phone, characterized in that the user can individually set the speed, intonation, volume according to taste. The method of claim 2, The web analysis unit, A command line analysis unit that analyzes user input information through the data input unit and establishes a suitable web information access system; A web access unit that connects to a website according to the instruction line analysis unit; A data receiver configured to store web document data that can be collected by HTTP information from a website accessed by the web access unit; A data inspecting unit for validating the web document data of the data receiving unit and processing normal / error according to the validity checking result; A data alignment unit for segmenting the web document data normally processed by the data inspection unit into individual units and rearranging the web document data according to one-dimensional information; A data filter unit for dividing data convertible into text information among the web document data rearranged by the data sorting unit, and requesting the web correction unit to query the data which is not interpreted as the text information; A data inquiry unit for converting web document data transmitted from the data filter unit into text information and requesting an analysis of a web image entity; The web image object delivered from the data retrieval unit combines organic information such as internal information such as a picture file name, tool tip information, caption information, a location of the web image, correlation with surrounding information, and lower page information connected to the web image. Web image analysis unit for interpreting the text information; And The data transmission unit which combines the web document data textified by the data inquiry unit and the web image analysis unit, and transmits the text information arranged in accordance with the one-dimensional information to the data output unit Voice portal service providing system using a phone comprising a. The method of claim 2, The data output unit is interlocked with the speech synthesis unit to provide a user interface system, to accommodate user input information of the data input unit, and to be designed as a menu type IVR structure that can accommodate a large amount of command systems. Voice portal service providing system using a telephone. delete delete delete delete

Images