KR20110030947A - Internet protocol television broadcasting system, server and apparatus for generating lexicon - Google Patents

Abstract

PURPOSE: An IPTV broadcasting system, a server and an apparatus for generating a lexicon are provided to offer a lexicon for voice recognition of the IPTV. CONSTITUTION: A pattern database(111) stores a voice generation pattern in order to offer an IPTV(Internet Protocol Television) broadcast service through the voice recognition. Using the voice generation pattern, a lexicon generating unit(110) generates a lexicon for the voice recognition corresponding to a voice command of a user. A structure information extraction unit extracts the extraction type keyword from the IPTV broadcasting information data.

Description

IPTV Broadcasting System, Server and Apparatus for Generating Lexicon {Internet Protocol Television Broadcasting System, Server and Apparatus for Generating Lexicon}

The present invention relates to an IPTV broadcasting service, and more particularly, to an IPTV broadcasting system, a server, and a voice list generating apparatus capable of recognizing a user's voice and providing an IPTV broadcasting service corresponding thereto.

The present invention is derived from a study conducted as part of the IT growth engine core technology development project of the Ministry of Knowledge Economy. [Task management number: 2006-S-036-04, Task name: Large-capacity interactive distributed processing voice interface technology for the new growth engine industry] Development].

With the development of digital technology, broadcasting services through the Broadband Convergence Network (BcN) are increasing. In particular, over the air, wired, terrestrial, film, music, and two-way systems are provided through a TV connected to an IP-based communication network. IPTV broadcasting services that can provide one-way / two-way services such as quiz shows, TV banking, and Internet search are attracting attention.

In an IPTV broadcasting service, a provider transmits content of a channel selected by a user through a head-end, and an indoor user broadcasts an IPTV through a set-top box (STB) and a TV connected thereto. You can get services.

Conventional TVs provide a remote control only for the convenience of user operation, but IPTV can recognize and manipulate a user's voice to select a specific menu, input a specific command, request a specific channel, or select a content by incorporating voice recognition technology. It provides convenience.

Such a speech recognition system applies the isolated word recognition method and the continuous word recognition method according to the pronunciation form.

First, the isolated word recognition method used in mobile phone voice dialing or the like is read by cutting off each word, and there is a considerable silence section before and after the word, so it is easy to grasp the beginning and end of the word and has a high recognition rate.

In addition, the continuous word recognition method recognizes speech in units of sentence, recognizes normal speech sentences, and each sentence does not add silence between words. Therefore, the continuous speech recognition method is somewhat difficult to recognize speech due to a coarticulation effect in which the characteristics of one word are affected by the pronunciation of adjacent words.

Therefore, most domestic and foreign speech recognition systems adopt the isolated word recognition method with high recognition performance within a limited application range. However, the isolated word recognition method needs to prepare a vocabulary to be recognized in advance to recognize the isolated word in advance, so a method for effectively preparing the isolated word is required.

In order to solve the above problems, an object of the present invention is to provide an IPTV broadcasting system, a server and a voice list generating apparatus capable of generating a voice list of voice words for voice recognition according to various voice types.

In accordance with an aspect of the present invention, an apparatus for generating a voice list of an IPTV broadcast includes: a pattern database configured to store one or more voice word generation patterns for providing an IPTV broadcast service through voice recognition; And a utterance list generator configured to generate a utterance list of voice recognition utterances corresponding to a user's voice command using the utterance generation pattern.

An IPTV broadcasting system according to another aspect of the present invention is an IPTV broadcasting server for generating and providing a speech list including one or more speech words corresponding to a user's voice command to provide an IPTV broadcasting service according to a user voice command. ; And an IPTV set-top device that recognizes a user voice command and interprets and processes the voice command recognized using the speech list.

According to another aspect of the present invention, an IPTV broadcast server includes: a first database for storing IPTV broadcast data; A voice list generator for generating a voice list corresponding to a user command using the IPTV broadcast data; A second database for storing the generated voice list; A command interpreter that interprets the recognized speech using the speech list; A command processor for searching IPTV broadcast data corresponding to the interpreted voice or executing an application of the IPTV broadcast corresponding to the interpreted voice; And a transmitter for transmitting the searched IPTV broadcast data or an execution result of the application to an indoor device.

According to the present invention, by categorizing and defining the voice word types in detail and generating voice words in consideration of each type, there is an effect of providing a voice list for IPTV voice recognition corresponding to various voices of the user.

As described above, recent IPTV broadcasts provide not only a touch interface such as a keyboard but also an interface for voice recognition. In order to provide an effective voice recognition interface, the present invention extracts and classifies utterances related to uttered words from broadcasting information data of IPTV broadcasting, which is produced on a text basis, to generate and provide utterance lists corresponding to various utterances of users. .

For the sake of understanding, prior to describing the configuration of the present invention, a spoken word constituting a speech list for various speech recognitions generated by the present invention will be described.

According to the contents, there are device operation commands, menu operation commands, content keywords, single domain main control, complex domain main control, and restricted domain main control, and according to the format, there are single voice and natural language voice. Hereinafter, each uttered word will be briefly described.

Device operation commands are commands for manipulating IPTV set-top devices, such as 'Turn On', 'Power Off', 'Volume Up', and 'Volume Down'.

The menu operation command is an application execution command that performs menu switching or content request by manipulating a user interface (UI) of the IPTV broadcasting service. For example, the menu is 'top menu', 'my home', and 'latest movie order'. And the lowest price.

Content keywords are classified into dictionary keywords and extracted keywords, and are strings extracted from IPTV broadcasting information data describing information of IPTV broadcasting contents.

First, a dictionary keyword is a word composed of a single noun extracted from IPTV broadcasting information data. For example, a proper noun representing a person's name, place name, organization name, etc., such as' Lee Hyo-ri ',' Choi ',' Samsung COEX ', or' HIV ( AIDS) 'and' Writing Substantive Examination '.

Next, the extracted key words are the main title or additional information (e.g., model name, episode information, trailer, etc.) extracted from the IPTV broadcasting information data. For example, if the IPTV broadcasting information data is "<okay, don't cry> trailer" The key words may be 'do not cry alright' or 'trailer', and in the case of "12th VJ Network-Achasan, Icheon Hanwoo Farm, Chungju-si," the title may be 'VJ Network' and the round information '12times'. Canon DSLR EOS-500D "may be the model name 'EOS-500D'.

Single domain main control (or domain main control) is not only VOD content, but also frequently used in information services such as shopping or weather, classification words used for classification, terms describing features, terms frequently used in specific areas, etc. to be. For example, in the shopping area, taxonomy such as 'apparatus', 'furniture', 'book', 'office supplies', genres such as 'action', 'drama', 'SF', etc. Temperature ',' minimum temperature ',' humidity 'and' precipitation '.

Complex domain control consists of a combination of two or more single domain control, or a combination of two or more nouns and a single domain control, such as 'Fantasy Movie', 'Thriller Movie', or 'Next Week', 'Tomorrow Stock Price'. Trends, 'today's screening program.

Constrained domain control is used to search for content and uses a qualifier to limit the search target. That is, it is composed of a combination of modifiers and domain principal control, and the modifiers may be composed of tubular phrases (eg, verbs + tubular endings, etc.). For example, it may be a family-friendly movie, a movie recently released, a nearest subway station, and the cheapest badminton racket.

Single word utterances are not words, but words that end with nouns. Natural words are words that consist of sentences that are predicates (verb + final ending).

As described above, the present invention relates to a voice list generating apparatus capable of effectively generating a voice list by combining various types of voice words, an IPTV broadcast server and an IPTV broadcast system to which the voice list generating apparatus is applied. Hereinafter, with reference to the drawings will be described in more detail with respect to the technical features of the present invention.

Hereinafter, an apparatus for generating a voice list of an IPTV broadcast according to an embodiment of the present invention will be described with reference to FIG. 1. 1 is a block diagram showing an apparatus for generating a voice list of an IPTV broadcast according to an embodiment of the present invention.

As shown in FIG. 1, an apparatus 240 for generating a voice list of an IPTV broadcast includes a voice word generation pattern database 111 and a voice list generator 110.

The speech generating pattern database 111 stores one or more speech generating patterns for generating a speech recognition speech list for providing an IPTV broadcasting service through speech recognition. Here, each utterance generation pattern may be determined in consideration of various kinds of utterance words, that is, a device operation command, a menu operation command, a content key word, a single area main control, a complex area main control, or a restricted area main control.

Here, when a new service is added to the IPTV broadcast service to provide an IPTV broadcast service through voice recognition, the spoken word generation pattern database 111 is updated to further include a voice word generation pattern of the new service.

The voice list generating unit 110 applies one or more voice words to the voice word generation pattern stored in the voice word generation pattern database 111 to generate a plurality of voice lists corresponding to voice commands of the user recognized from the IPTV broadcasting information data. Create That is, the voice list generator 110 may generate one voice list corresponding to various voice commands of a user who wants to search IPTV broadcast data by including one or more voice words according to the voice word generation pattern.

Here, the user's voice command relates to a device operation command, a menu operation command, a content keyword, a single area main control, a complex area main control, a restricted area main control, and the like.

Hereinafter, a voice word generation pattern stored in the voice word generation pattern database 111 will be described with reference to Table 1 below. Here, Table 1 below is a speech generation pattern defined by the BNF (Backus Naur Form) syntax.

The utterance pattern is composed of one or more Term as shown in Table 1, and the Term may be any one of a nonterminal term, a terminal term, and a variable (b).

Here, the nonterminal term is a file name in which another utterance generation pattern existing between a start character "<" and an end character ">" is described (c), and the terminal term is a start character "" ", It may be a utterance string existing between the terminating character "" ". In addition, the variable may be a string consisting of the content key word extracted from the IPTV broadcasting information data (e), and the iterators are '+', '*' and '?' Included after the Term, and '+' is the same Term as 1 '*' Means that the same Term is omitted or can be repeated one or more times, and '?' Means that the same Term is omitted or used one time (f).

Hereinafter, an example of a utterance generation pattern that can be actually used will be described with reference to Table 2 below. Table 2 is an example of a utterance generation pattern according to an embodiment of the present invention.

The utterance generation patterns in Table 2 are the same as the contents of 'movie utterance list.txt', respectively, and refer to 'actor name.txt', 'genre name.txt', and content keywords extracted from IPTV broadcasting information data.

The utterance list generation unit 110 may generate a utterance list as shown in Table 3 below by using the utterance generation pattern described in 'movie utterance list.txt'.

VJ Network 12 times in Table 3, VJ Network is the content key words extracted from the IPTV broadcasting information data, "12 times VJ Network-Achasan, Icheon Hanwoo Farm" in Table 4, line 4 $ heading $ 5 times, $ 5 heading It's mapped. That is, the vocalization list generating unit 110 may generate a vocalization list as shown in Table 3 by mapping the main title 'VJ network' and the circumference '12 times' to (C) and (K).

The voice list generating unit 110 includes one or more voice words such as a device operation command, a menu operation command, a single area main control, a complex area main control, or a restricted area main control in the manner illustrated by Tables 1 to 3 You can create a list.

On the other hand, the IPTV broadcasting utterance list generating device 240 and the structure information extraction unit 120 for extracting the extracted key words and dictionary keywords from the IPTV broadcast information data, respectively, when the user's voice command is related to the content key words, The structure information extracting unit 130 and a extracted speech normalization unit 140 for processing the extracted core words into natural pronunciation words that are close to the user's pronunciation and providing them to the speech list generation unit 110 are provided. Hereinafter, each part is demonstrated.

The structural information extracting unit 120 uses a regular expression or a pattern rule from the IPTV broadcasting information data, for example, when a predetermined symbol is included in the IPTV broadcasting information data, or in a markup language format such as HTML. The rule 121 is identified, and the title or additional information is extracted from the IPTV broadcasting information data using the identified rule.

For example, when the IPTV broadcasting information data for the VOD content is "<okay, don't cry> trailer", the structure information extracting unit 120 grasps the rule 121 of the symbol characters '<' and '>'. You can extract the title 'Don't cry alright' and the additional information 'trailer'.

The unstructured information extraction unit 130 extracts proper nouns or terminology included in the IPTV broadcasting information data. That is, the unstructured information extracting unit 130 checks whether a character string included in the IPTV broadcasting information data exists from the dictionary 131 including proper nouns or terminology, and extracts it if it exists. However, since the IPTV broadcasting information data may have a person name or a character name which is not registered in general, such as "EBS fart blacksmith", if the broadcasting information data of the content includes a proper noun that is not registered, it is not registered. A procedure for registering the proper noun in the dictionary 131 by the manager of the voice list generation unit 110, the IPTV broadcasting service related person, the content creator, and the like may be required.

The speech normalization unit 140 uses the user's voice command based on a predetermined rule to extract the main title, additional information, proper nouns, and terminology extracted by the structural information extractor 120 and the non-structural information extractor 130. Convert to a form close to.

For example, the utterance normalization unit 140 converts 'MR' extracted from the IPTV broadcasting information data of "MR. Huayu" into "Mister" close to the user's voice command and converts the IPTV broadcasting of "Kim Kwan-jang VS Kim Kwan-jang VS Kim Kwan-jang". Converts 'VS' extracted from information data to 'large', converts 'ST' extracted from IPTV broadcasting information data called "ST. ELMO'S FILE" to "St." The '9 end 2 out' extracted from the information data can be converted to the 'end to end out'.

Hereinafter, an IPTV broadcasting system to which an apparatus 240 for generating a voice list of IPTV broadcasting according to an exemplary embodiment of the present invention is applied will be described with reference to FIG. 2. 2 is a block diagram showing an IPTV broadcasting system according to an embodiment of the present invention.

As shown in FIG. 2, the IPTV broadcasting system 20 according to the exemplary embodiment of the present invention includes an IPTV broadcasting server 200 and an IPTV set-top device 300.

The IPTV broadcasting server 200 generates and provides a speech list including one or more speech words corresponding to a user's voice command to provide an IPTV broadcasting service according to the recognized voice, and includes a broadcasting data database 211, The information data database 212, the manager 250, the voice list generator 240, the command interpreter 230, the searcher 220, the application performer 260 and the transmitter 270 are included. Hereinafter, each component of the IPTV broadcasting server 200 will be described.

The broadcast data database 211 stores IPTV broadcast data (eg, content and VOD files, etc.) for the IPTV broadcast service, and the information data database 212 stores IPTV broadcast information data (eg, a title of the content, etc.). .

The management unit 250 stores and manages IPTV broadcast data and broadcast information data, and updates the broadcast data database 211 and the information data database 212 when a new service or content is added to the IPTV broadcast. At this time, the management unit 250 may manage the IPTV broadcast data and broadcast information data by dividing the fields into storage date, time, content type, source, producer information, and the like.

The voice list generator 240 generates a voice list by including one or more voice words to provide an IPTV broadcasting service according to the recognized voice, and stores the generated voice list in the voice list database 241. By the way, broadcast data is constantly updated, but since the storage space of the utterance list database 241 is limited, the utterance list generation unit 240 (or the management unit 250) stores the utterance list in association with the broadcast information data. By doing so, it is possible to perform management such as deleting an unused talk list of the talk list database 241.

The voice list generating unit 240 may generate a voice list through the method of Tables 1 to 3 described above, and transmits the generated voice list to the IPTV set-top device 300 in real time or periodically. To recognize a new IPTV broadcast.

The command interpreter 230 interprets the recognized user voice command using one or more voice lists. That is, the command interpreter 230 receives and interprets a character string corresponding to a user voice command from the IPTV set-top device 300, and transmits the interpreted command to the searcher 220 if the interpreted command is a command for requesting IPTV broadcast data search. If a command for requesting the execution of the application of the IPTV broadcast is transmitted to the application execution unit 260. Here, the application may be related to internet search, home shopping search, purchase, weather inquiry, stock inquiry, stock inquiry, and the like.

The search unit 220 receives a command for requesting IPTV broadcast data search from the command interpreter 230 and searches for IPTV broadcast data corresponding to the voice command interpreted from the broadcast data database 211.

The application execution unit 260 receives a command for requesting the execution of the application of the IPTV broadcast from the command interpreter 230, and executes the application of the IPTV broadcast corresponding to the interpreted command. In detail, the application execution unit 260 executes a plurality of applications stored in the application database 261, and transmits the application execution result to the transmission unit 270.

For example, if the received string is the "cheapest badminton racket", the application execution unit 260 makes an SQL statement to sort badminton racket products among the shopping data stored in the application database 261, and the lowest price badminton racket therefrom. The search may be provided to the IPTV set-top device 300.

Or, if the received string is "Today's weather in Daejeon region", the current date and the Daejeon region are entered in the date field and the region field, respectively, for the weather information stored in the application database 261 using an SQL statement, and the weather information is searched. To the IPTV set-top device 300.

The transmitter 270 transmits the IPTV broadcast data searched by the searcher 220 or a result (eg, a UI menu) of the application execution of the application execution unit 260 to the IPTV set-top device 300.

Meanwhile, the search unit 220 and the application execution unit 260 of the IPTV broadcasting server 200 may not be divided into separate components, but may be merged into a processing unit (not shown) that performs data search or application execution. Of course. Hereinafter, the IPTV set top apparatus 300 will be described.

The IPTV set-top device 300 recognizes a user's voice command and interprets and processes the recognized voice command using the one or more voice lists, and the voice list database 310, the voice recognizer 320, and the command interpreter 330, a controller 340, and a display 350.

The voice list database 310 receives and stores one or more voice lists from the IPTV broadcasting server 200. At this time, the voice list database 310 receives a voice list for a new IPTV broadcast or a service thereof from the IPTV broadcast server 200 in real time or periodically.

The voice recognition unit 320 recognizes a user voice command and converts the voice command into a string using one or more voice lists.

The command interpreter 330 checks whether the string received from the voice recognition unit 320 is a device manipulation command, and requests an IPTV broadcasting service corresponding to the text string if it is not a device manipulation command. Then, the command interpreter 330 of the IPTV broadcast server 200 receives a character string, and interprets whether the command is a command for requesting IPTV broadcast data or a request for execution of an application of IPTV broadcast, and searches according to the analysis result. The interpreted command is transmitted to the unit 220 or the application execution unit 260.

When the controller 340 receives the device operation command from the command interpreter 330, the controller 340 controls the operation of the IPTV set-top device 300 accordingly. At this time, the device operation command is a command for operating the IPTV set-top device as described above, for example, 'Turn On', 'Power Off', 'Volume Up', 'Volume Down 4', and the like.

The display unit 350 retrieves the IPTV broadcast service provided according to a request of the command interpreter 330 or provided according to the execution of the application. In this way, the present invention can provide an IPTV broadcasting service according to voice recognition.

While the present invention has been described in detail with reference to the accompanying drawings, it is to be understood that the invention is not limited to the above-described embodiments. Those skilled in the art will appreciate that various modifications, Of course, this is possible. Accordingly, the scope of protection of the present invention should not be limited to the above-described embodiments, but should be determined by the description of the following claims.

1 is a block diagram showing an apparatus for generating a voice list of an IPTV broadcast according to an embodiment of the present invention.

2 is a block diagram showing an IPTV broadcasting system according to an embodiment of the present invention.

Claims (10)

A pattern database for storing one or more voice word generation patterns for providing an Internet Protocol Television (IPTV) broadcasting service through voice recognition; And A voice list generator for generating a voice list of voice recognition voice words corresponding to a user's voice command using the voice word generation pattern Voice list generating apparatus of the IPTV broadcast comprising a. The method of claim 1, The utterance list includes one or more utterances including content keywords, device operation commands, menu operation commands, single area control, complex area control, or restricted area control. The content keyword is a voice list generating apparatus of the IPTV broadcast, which is composed of a character string included in the IPTV broadcast information data. The method of claim 1, A structure information extracting unit which extracts an extractable key word from the IPTV broadcasting information data; Unstructured information extraction unit for extracting a dictionary key word from the IPTV broadcast information data; And A speech normalization unit which processes the extracted core words into natural pronunciation words that are close to the user's pronunciation and provides them to the speech list generation unit. Voice list generation device of the IPTV broadcast further comprising a. The method of claim 1, wherein the pattern database, When a new service is added to the IPTV broadcasting service, the voice list generating apparatus of the IPTV broadcasting, which is updated to further include a voice word generation pattern for the new service. An IPTV broadcasting server for generating and providing a speech list including one or more speech words corresponding to a user's voice command to provide an Internet Protocol Television (IPTV) broadcasting service according to a user voice command; And An IPTV set-top device that recognizes user voice commands and interprets and processes the voice commands recognized using the speech list. IPTV broadcasting system comprising a. The method of claim 5, wherein the IPTV set-top device, A voice list database for receiving and storing the voice list; A voice recognition unit recognizing the user voice command and converting the voice command into a character string using the speech list; A command interpreter which checks whether the character string is a device operation command and requests an IPTV broadcast service corresponding to the character string to the IPTV broadcast server when the character string is not a device operation command; A control unit controlling an operation of the IPTV set-top device according to the device operation command; And Display unit for displaying the IPTV broadcast service provided according to the request IPTV broadcasting system comprising a. The method of claim 5, wherein the IPTV broadcast server, A broadcast data database for storing IPTV broadcast data; A voice list generator for generating a voice list including the one or more voice words to provide an IPTV broadcasting service according to the recognized voice; A command interpreter that interprets the recognized voice command using the speech list; A search unit for searching IPTV broadcast data corresponding to the interpreted voice command from the database; An execution unit executing an application of an IPTV broadcast corresponding to the interpreted voice command; And Transmitter for transmitting the searched IPTV broadcast data or the results of the application execution to the IPTV set-top device IPTV broadcasting system comprising a. A first database for storing IPTV broadcast data; A voice list generator for generating a voice list corresponding to a user command using the IPTV broadcast data; A second database for storing the generated voice list; A command interpreter that interprets the recognized speech using the speech list; A command processor for searching IPTV broadcast data corresponding to the interpreted voice or executing an application of the IPTV broadcast corresponding to the interpreted voice; And Transmission unit for transmitting the searched IPTV broadcast data or the execution result of the application to the indoor device IPTV broadcast server comprising a. The method of claim 8, wherein the voice list generating unit, And generating the utterance list using a utterance generation pattern including at least one of unstructured data and structured data for IPTV broadcasting. The method of claim 8, wherein the second database, And stored in association with the first database.

Images