KR100681765B1 - Web and Telephony Network-Interactive Study System Using VoiceXML - Google Patents

Abstract

The present invention relates to a web and telephone network interworking learning system using VoiceXML technology implemented to enable a user to learn according to their purpose using a telephone network and the web. Learning system of the present invention, the telephone network interface means for the user output through the decoding and voice synthesis of the voice command of the learner using the telephone network, the Internet network access means for accessing the internnet network, the database means for storing the learning content And a learning means for controlling user ID and authentication process, and learning management means for storing learning progress and suggesting an effective learning schedule. A voice communication terminal of a user connectable to the telephone network interface of the learning system apparatus through a public telephone network and a mobile communication telephone network; And a computer terminal of a user connectable to the internet network access means of the learning system apparatus through an internet network.

Learning System, Voice XML, Telephone Network, Internet

Description

Web and Telephony Network-Interactive Study System Using VoiceXML}             

1 is a block diagram of a learning system device proposed in the present invention.

2 is a diagram illustrating a web flow diagram of the entire learning system proposed in the present invention.

3 is a diagram illustrating a voice XML flow chart of the entire learning system proposed by the present invention.

4 is a view showing a user registration and user authentication flow of the learning system proposed in the present invention.

5 is a diagram illustrating an example of a system configuration of a learning system proposed by the present invention.

6 is a view showing the flow of the algorithm for registering the user ID, password, name using the voice proposed in the present invention.

7 is a view showing another example of the system configuration for the learning system proposed in the present invention.

<Explanation of symbols for the main parts of the drawings>

100: learning system device 110: telephone network interface unit

120: Internet network access unit 130: database unit

140: control unit 150: learning management unit

500: Voice XML Gateway 500A: PSTN Interface

500B: Speech Recognition Engine 500C: Speech Synthesis Engine

510 Web Server 510A Voice XML Document

510B: Voice XML Database 520: Mobile Phone Terminal

530: base station 540: general telephone terminal

550: computer terminal

The present invention relates to a learning system, and more particularly, to a web and telephone network interworking learning system using VoiceXML technology implemented to enable a user to learn according to their purpose using a telephone network and the web. It is about.

In recent years, the implementation goal of the learning system is commonly used in the term of distance education, and the goal is to improve the quality of teaching, facilitate access to education, and reduce the cost of education. In addition, in the case of the existing learning system, distance education was adopted as a way to restrain the cost of education, but this was mostly implemented through one method such as telephone or internet. However, such a learning system was a convenience-oriented implementation that did not consider the learner's situation, and it was an inappropriate construction environment to increase the efficiency of education.

In other words, the learning system using the telephone network has the advantage of using a tool that is easy to carry anytime and anywhere, but it was difficult to conduct an in-depth inquiry compared to the learning using a computer.

In addition, the computer learning system using the Internet is capable of in-depth learning compared to the learning system using the telephone, but has a disadvantage in that the learner should always carry a computer tool in a fixed place.

Therefore, in order to make up for the shortcomings of the existing learning system, it is necessary to implement a learning system in which a web and a telephone network are interworked, and to implement a learning system that can be used by learners anytime and anywhere.

The present invention has been made in order to meet the above-mentioned demands of the prior art, by implementing the user ID registration function using the web and telephone network, the individual user progress management function, it is possible to evaluate the learning results for measuring the learning effect, The purpose is to provide a web and telephone network interworking learning system using VoiceXML technology implemented so that users can learn according to their purpose.                         

Learning system of the present invention for achieving the above object, the telephone network interface means for the user output through the decoding and voice synthesis of the voice command of the learner using the telephone network, Internet network access means for accessing the internnet network, and learning A learning system apparatus including database means for storing contents, control means for controlling user ID and authentication process, and learning management means for storing learning progress and suggesting an effective learning schedule; A voice communication terminal of a user connectable to the telephone network interface of the learning system apparatus through a public telephone network and a mobile communication telephone network; And a computer terminal of a user connectable to the internet network access means of the learning system apparatus through an internet network.

Another learning system of the present invention for achieving the above object comprises: a web server connected to the Internet network and constructing a learning voice XML document and a user database to manage user learning; Connected to a telephone network and connected to the web server through the Internet, the telephone network interface unit receiving a call of a user who wants to use a learning system using the telephone network, and detecting and detecting a voice signal section via the telephone network interface unit. And a speech recognition engine unit for reading a voice signal and recognizing a user command, and a speech synthesis engine unit for outputting the result after the user command processing of the control unit as a voice to be heard by the user. A voice XLM gateway configured to voice synthesize a training voice XML document and deliver the voice voice document to a user through a telephone network; A voice communication terminal of a user connectable to the voice XLM gateway through a public telephone network and a mobile communication network; It is configured to include a computer terminal of a user that can access the web server through the Internet network.

Here, when a user via a telephone network and an internet network performs an authentication function and a learning function, the same learning progress is stored and personalized information is provided even for changing use of the telephone network or the internet network by utilizing and storing information of the same database.

In order to increase the efficiency of speech recognition performance when performing registration function of ID, password and name using voice through the telephone network, segment recognition of initial, neutral, and final syllables per syllable is performed without performing recognition function by syllable or word unit. After creating a syllable function through the combination.

After performing the registration function of user ID, password and name, when re-login, ID, password, and name are used, the pre-stored information is used to perform the recognition unit in word units.

Hereinafter, a web and telephone network interworking learning method and system using voice XML according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a learning system device proposed in the present invention.

Learning system apparatus according to the present invention, the telephone network interface unit 110 for the user output through the decoding and speech synthesis of the voice command of the learner using the telephone network, the Internet network connection unit 120 for accessing the internnet network, and learning Database unit 130 for storing the content, the control unit 140 for controlling the user ID and authentication process, and the learning management unit 150 for storing the learning progress and propose an efficient learning schedule is configured to include .

The telephone network interface 110, the telephone network interface unit for receiving a call of the user who wants to use the learning system using the telephone network, and the detection of the voice signal interval via the telephone network interface unit and read the detected voice signal and the user command And a speech recognition engine unit for recognizing the voice recognition engine unit for outputting the result after the user command processing of the control unit 140 to be heard by the user. The telephone network interface 110 performs a voice command recognition and synthesis function using the VXML technology.

The internet network access unit 120 is used for accessing the internet network of the learning system, and is responsible for transmitting and receiving commands and control of the learning system user using the internet network by using wired or wireless interfaces such as Lan, Wireless Lan, and cable modem. do.

The database unit 130 is an essential part of the learning system according to the present invention, and is responsible for storing learning contents, storing user ID and password, and storing personal progress. In addition, unlike the existing learning system, the database unit 130 is commonly used when operating the learning system according to the user command via the VXML Interpreter via the telephone network interface unit 110 and the Internet network access unit 120. Has a structure. 2 and 3, through the sharing of the DB unit 130 (DB1 and DB2 in FIG. 2 is the same database unit), the learning system of the present invention is the progress of the user of the learning system via the telephone network and the Internet network Functions such as storage, ID management and password management can be used. In addition, the technical problem of sharing a database on the web by voice using a telephone network can be solved through the application of VXML technology.

The controller 140 is a part for processing a user command received from the telephone network interface unit 110 and the internet network access unit 120, and is responsible for controlling and operating the learning system of the present invention. The user command received by the control unit 140 determines the operation according to the user's intention, calls the information of the database unit 130 or the learning management unit 150, and calls and manages the learning data according to the user's individual progress. .

In addition, the control unit 140 of the present invention is responsible for the registration of the user ID and password registration function for the first use of the user learning system. In the existing learning system, the user ID and password registration function of the learning system using the Internet network could be easily implemented by exchanging text data. However, if you are a user of the learning system via the telephone network, you need to recognize and register all your IDs, passwords, names, etc. by using the voice recognition function. Recognition should be performed with 2,350 recognition candidates, such as the number of complete Hangul syllables. It is difficult to implement the performance of the speech recognition engine to perform the speech recognition function of the existing system, and it is difficult to practically use it as a cause of the decrease of the recognition rate.

Therefore, the present invention proposes to recognize the ID, password, user name and the like utilizing the algorithm shown in FIG. That is, in order to prevent the recognition rate deterioration caused by the recognition function for the 2,350 syllables, the following function is performed, for example. That is, in order to register his name, “Hong Gil-dong”, the user first selects three number of recognized syllables and inputs the number of syllables to be recognized by the learning system according to the guidance message of the learning system (step S600). It is checked whether the syllables have initial, neutral, and finality (step 610). In this example, the number of syllables has three syllables, and each syllable has all the consonants, the neutral and the finality. Therefore, the performance of the recognition function is the consonant group “a, ㄲ, b, c, ㄸ, d, ㅁ, ㅂ, ㅃ, ㅅ, ㅆ, ㅇ, ,, ㅉ, ,, ㅋ, ㅌ, ,, ㅎ ”are selected from 19 candidate groups (step S620), and the recognition function is performed by the voice of the candidate groups (step S620). S630). In other words, the user can say “memory” to input “a” as a consonant, and “neun” to enter “b”. In this case, a voice recognition function for 19 words may be performed. In the case of neutral, the same method can be applied to the recognition of 20 words even when considering both vowels and double vowels. It is possible to input all available Hangul alphabets. That is, even if all of the initial, neutral, and final species are combined, the number of recognition candidates is only 46, which may indicate a high result in recognition rate when compared with 2,350 recognition candidates. In addition, if the recognition process of initial, neutral, and finality is set separately for each page using VXML technology, the number of candidates for recognition becomes smaller.In the case of initial, 19, and 20, Only the recognition function for the seven words needs to be performed (step S630).

After performing the voice recognition function for the initial, neutral, and final, the syllable is implemented by combining the initial, neutral, and final in the learning system, and then confirming the recognition result by notifying the user through the telephone network. If it is confirmed that the process proceeds to the above-described step S620 and the following steps are performed again, and if it is confirmed by the recognition, it is checked whether the syllable additional recognition is necessary (step S660). The following steps are repeated.

On the other hand, if it is determined in step S660 that no additional recognition is required, an ID, a name, and a password are generated through the syllable combination input above. The user ID, password, and name generated as described above are documented and stored in the user DB, and the stored voice XML document can be used as an authentication candidate via voice or internet network when the user reconnects to use the learning system. do.

It is obvious that the ID, password, and user name registration algorithms proposed in the present invention can be applied not only to the present learning system but also to other types of systems for inputting Korean voices in voice applications.

The learning system manager 150 decodes the user command received from the telephone network interface 110 or the internet network access unit 120, and after completing the user authentication process of the controller 140, displays the individual learning information of the authenticated user. Transfer to the control unit 140. At this time, the personal learning information of the user is automatically updated in the user personal information column on the learning system manager 150 at the end of the learning system.

2 and 3 show a web and voice XML processing flowchart of the entire learning system proposed in the present invention.

The learning system constructed as shown in FIG. 1 is installed to enable access to the telephone network and the Internet network, and then the service according to the flow shown in FIGS. 2 and 3 when the learning system user accesses the telephone network or the Internet network. To provide.

First, when the user connects via the computer terminal via the Internet network, the guide information for the subject is presented to the user by using the computer (step S102), and it is determined whether the member is registered (step S104). In the case of a user attempting a new connection, a member registration function is performed (step S106), the member registration information is stored in DB1 (S108), a member registration message is guided, and confirmation is received (S110). If the existing member performs the login function through the authentication function (step S112), after confirming the member information in DB1 according to the login, and confirms the personal progress of the logged in member in DB1 (S114 ~ S116).

Thereafter, the user selects the learning unit through the computer terminal and selects the learning range (steps S118 to S220). When the outline of the learning range is selected according to the learning range, the outline description is output to the user's computer terminal (steps S122 to S124). When the term is selected, the description of the term is output to the user's computer terminal (step S126). ~ S128). In addition, if an O / X problem or a short answer problem is selected (steps S130 and S132), the question is asked to the user computer terminal about the problem and when the user submits an answer, the learning system checks the solution and guides the grades ( Steps S134 to S138).

Subsequent to steps S124, S128 and S138, the individual progress is stored in DB1 (S140), and if there is no learning end selection, the process proceeds to step S118.

On the other hand, the guide information for the subject when the user accesses via the telephone network through a general telephone or a mobile phone terminal is presented to the user through a general telephone or a mobile phone terminal (step S202), membership registration according to the member's voice input It is determined (step S204). In the case of a new user attempting to connect, a user registration function is performed according to a voice input (step S206), and the voice input is stored as membership information in DB2 (S208), and a voice prompt for a member registration message is confirmed. It is received (S210). If the existing member performs the login function through the authentication function according to the voice input (step S212), after confirming the member information in DB2 according to the login, and confirms the personal progress of the member logged in DB2 (S214 ~ S216).

Thereafter, the learning unit is selected by the user's voice input through the general telephone or the mobile phone terminal, and the learning range is selected (steps S218 to S220). When the outline of the learning range is selected by the user's voice input according to the corresponding learning range, the outline description is output to the user's general telephone or mobile phone terminal (steps S222 to S224), and when a term is selected according to the user's voice input The description of the term is output to the user's cellular phone terminal or a general telephone (steps S226 to S228). Also, if an O / X question or a short answer is selected by the user's voice input, the question is asked. If the answer is submitted by the user's voice, the answer is checked and the question is repeatedly performed until all questions are provided. After (steps S230, S232, S234), when all questions are submitted and the solution is confirmed, the learning system guides the grades to the user's general telephone or mobile terminal (step S236).

Subsequent to the above steps S224, S228 and S236 to store the individual progress in DB2 (S238), if the user is requested to connect to the agent after the call with the agent is made (steps S240, S242), the user's voice input If there is no learning end selection, the process proceeds to step S218.

4 is a diagram illustrating an example of a user registration and user authentication flow of the learning system proposed by the present invention.

When the user accesses the learning system through the telephone network, a user name registration guide message is output from the learning system (step S302), and accordingly, when the user pronounces the initial, neutral, and finality with respect to his or her name (S304 to S308). After combining the letters in the guide by the user spoken letters (S310 ~ S312). At this time, if the user checks whether the letter is pronounced by the user and confirms that it is not, the process returns to the above-described step S304, while if the user checks the pronunciation of the letter, the pronunciation of all the letters after the above-mentioned step S304 is completed. Repeat the step (S314 ~ S316). When it is confirmed that the input of the letters of all names is completed, the learning system combines the names and stores them in the user database (step S318), and proceeds to step S320 to end the control after guiding the entered name to the user.

5 is a diagram illustrating a practical configuration example of the learning system proposed by the present invention. In the figure, the voice XML gateway 500 and the web server 510 replace the configuration of FIG. 1 described above. The voice XML gateway 500 includes the PSTN interface 500A, the voice recognition engine 500B, and the voice synthesis. It is comprised including the engine 500C. The web server 510 includes a voice XML document 510A and a database 510B. In this system, the function of the control unit 140 in the configuration of FIG. 1 is shared between the voice XML gateway 500 and the web server 510, and the function of the learning management unit 150 is the web server 510. The function of the telephone network interface unit 110 is the voice XML gateway 500.

In such a configuration, when the user accesses the voice XML gateway through a base station and a public telephone network using the mobile phone terminal 520, the voice of the user is correctly recognized through the voice recognition engine 500B, and the recognized user information is recorded on the web. It is delivered to the server and authenticated with or stored in database 510B. In addition, the voice XML document 510A of the web server 510 corresponding to the learning material is synthesized by the voice synthesis engine 500C of the voice XML gateway 500 by the user's request according to the user's voice. Delivered.

On the other hand, when a user connects to a web server through the Internet using a computer terminal, processing such as user authentication is processed by a general web method, and a voice XML document of the web server 510 corresponding to learning material by a user request is processed. 510A is also transmitted to the user computer terminal via the Internet.

7 is a diagram illustrating another practical configuration example of the learning system proposed by the present invention. In the example of FIG. 1, the learning system device of FIG. 1 is connected to a telephone network and an internet network, and the detailed description thereof will be omitted since it can be easily understood through the description of FIG. 1.

On the other hand, the present invention is not limited to the specific embodiments described above, but can be carried out by variously modified and modified within the scope not departing from the gist of the present invention.

According to the present invention, it is possible to selectively use the learning system using the telephone network and the Internet network according to the user environment, and the user can automatically save the learning progress when the learning system using the telephone network or the internet network is used. It is possible to use the learning system by utilizing the convenient method according to the method.

Claims (8)

Telephone network interface means for user's output through voice synthesis and speech synthesis of learner using telephone network, Internet network access means for accessing internnet network, database means for storing learning contents, user ID and authentication process A learning system device comprising a control means for controlling the learning and learning management means for storing the learning progress and an efficient learning schedule; A voice communication terminal of a user connectable to the telephone network interface of the learning system apparatus through a public telephone network and a mobile communication telephone network; A computer terminal of a user connectable to the internet network access means of the learning system apparatus through an internet network; In order to increase the efficiency of speech recognition performance when performing the registration function of ID, password and name using voice through the telephone network, the speech division and the speech division of speech, initial, neutral, and final syllables are not performed. Web and telephone network interworking learning method system using Voice XML technology characterized by generating a syllable function through the combination after performing the recognition. The method of claim 1, When the user via the telephone network and the internet network performs authentication and learning functions, the same database can be used for storing and personalizing the same learning progress even when the telephone network or the internet network is changed and used. Web and telephone network interworking learning method system using Voice XML technology. delete The method of claim 1, After performing the registration function of the user ID, password, and name, when re-login ID, password, name is recognized Voice X M (VoiceXML), characterized in that by using the pre-stored information, performing the unit of recognition by word unit Web and telephone network interworking learning method system using technology. A web server connected to the Internet network and constructing a learning voice XML document and a user database to manage user learning; Connected to a telephone network and connected to the web server through the Internet, the telephone network interface unit receiving a call of a user who wants to use a learning system using the telephone network, and detecting and detecting a voice signal section via the telephone network interface unit. And a speech recognition engine unit for reading a voice signal and recognizing a user command, and a speech synthesis engine unit for outputting the result after the user command processing of the control unit as a voice to be heard by the user. A voice XLM gateway configured to voice synthesize a training voice XML document and deliver the voice voice document to a user through a telephone network; A voice communication terminal of a user connectable to the voice XLM gateway through a public telephone network and a mobile communication network; A computer terminal of a user connectable to the web server through an internet network; In order to increase the efficiency of speech recognition performance when performing the registration function of ID, password and name using voice through the telephone network, the speech division and the speech division of speech, initial, neutral, and final syllables are not performed. Web and telephone network interworking learning method system using Voice XML technology characterized in that to generate a syllable function through the combination after performing the recognition. The method of claim 5, When the user via the telephone network and the internet network performs authentication and learning function, the same database can be used for storing and personalizing the learning progress even when the telephone network or the internet network is changed and used by storing and using the information of the same database. Web and telephone network interworking learning method system using VoiceXML technology. delete The method of claim 5, Voice XML technology, characterized in that the unit of recognition is performed in units of words by using the pre-stored information when recognizing the ID, password or name after performing the registration function of the user ID, password and name. Web and phone network interworking learning method system using the system.

Images