KR101700819B1 - Apparatus and method for speech recognition - Google Patents

Abstract

The present invention relates to a device for speech recognition and automatic interpretation operating on a PC or a mobile device, and a speech recognition device according to the present invention includes a screen for selecting a domain as a unit for a speech recognition area of a predetermined classification for speech recognition To the user; A user input unit for receiving a selection of a domain from the user; And a communication unit for transmitting the selection information of the user to the domain. According to the present invention, it is possible to provide a user with an intuitive and easy-to-use speech recognition device, which enables a user to easily select / modify a designated domain of a speech recognition system, Performance can be improved.

Description

 [0001] Apparatus and method for speech recognition [0002]

Field of the Invention [0002] The present invention relates to a device or a speech recognition method equipped with a speech recognition and automatic interpretation function, and more particularly to a method of selecting a domain of a database for speech recognition.

Conventional speech recognition or automatic interpretation systems are usually trained in only one domain, domain, since it is inefficient to train many vocabularies or expressions in various ways. Conventional speech recognition or automatic interpretation applications can not, in most cases, modify a domain that is designated by default. Further, even if the user can directly select the domain, there is a problem that the user is inconvenient to use and the selection content is very simple. Therefore, the adaptability to the speech recognition environment is lowered, and the accuracy of voice recognition and automatic interpretation is lowered.

The present invention provides users with a database to be referred to in speech recognition or automatic interpretation, that is, a user interface for easily selecting a domain, thereby making it easy to select a domain according to a situation, thereby improving the accuracy of speech recognition and automatic interpretation .

According to an aspect of the present invention, there is provided a speech recognition apparatus comprising: a display unit for displaying a screen for selecting a domain as a unit for a speech recognition area of a predetermined classification for speech recognition; A user input unit for receiving a selection of a domain from the user; And a communication unit for transmitting the selection information of the user to the domain.

According to the present invention, it is possible to provide an intuitive and simple domain selection method to a user, thereby improving the accuracy and performance of voice recognition and automatic interpretation.

1 is a network diagram for providing a speech recognition service according to an embodiment of the present invention.
2 shows a configuration of a user terminal according to an embodiment of the present invention.
3 illustrates an exemplary structure and relationship of a designated domain according to an embodiment of the present invention.
4 to 11 illustrate exemplary screens displayed on a display unit of a user terminal according to an exemplary embodiment of the present invention.
12 is a flowchart illustrating a method of specifying a speech recognition domain according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

1 is a network configuration diagram for providing a speech recognition method according to an embodiment of the present invention.

The user terminal 10 receives voice and domain selection information from a user and delivers the voice and domain selection information to the voice recognition server 20. The user terminal 10 may be any computing device equipped with a communication function such as a PC, a notebook, a smart phone, etc., and a user can input voice or text.

The speech recognition server 20 performs speech recognition by referring to data corresponding to a domain selected by the user among reference data for speech recognition stored in the DB 30 through the received speech and information on the selected domain. Then, the voice recognition result is transmitted to the user terminal 10.

The DB 30 stores various data necessary for speech recognition by the speech recognition server 20. Data such as a corpus, a language dictionary, and the like to be referred to during speech recognition are stored for each domain.

The user terminal 10 will now be described in more detail with reference to FIG.

2, the user terminal 10 according to the present embodiment may include a display unit 100, a user input unit 200, and a communication unit 300.

The display unit 100 displays information necessary for speech recognition, and may display menus for specifying a domain to be referred to by the user for voice recognition. In the present embodiment, the speech recognition server 20 receives a speech signal and recognizes the meaning of the speech signal, and performs speech recognition based on a designated domain or a general domain specified by the user.

A general domain is a database referred to in order to support speech recognition for a commonly used language other than a specific domain. A designated domain is automatically or automatically assigned to a specific situation to support speech recognition more accurately than the above- Is the database selected by the user. For example, if the input voice is related to travel, voice recognition can be performed using the 'travel' domain as the designated domain, and a voice recognition result can be generated that is better than when the general domain is selected.

The concept of a speech recognition domain is described in more detail with reference to FIG. In this embodiment, the speech recognition domain is a unit for classifying the speech recognition area, that is, a database to be referred to during the speech recognition process.

Referring to FIG. 3, as described above, the speech recognition server 20 can operate on the general domain 31 by default or by a user's selection. Here, each of the designated domains may have a first sub-domain 32 and a second sub-domain 33 having a first sub-domain 32 as a parent domain. Further, although not shown, it may include a third, fourth sub-domain having a second sub-domain as a parent domain.

These second sub-domains may replace some characteristics (words or expressions) of the parent domain, or may have characteristics that are not in the parent domain. Also, each of the domains may overlap with each other. For example, two subdomains, e.g., a Touring domain may partially overlap with a Business domain that is another domain, and a portion of a restaurant domain that is a subdomain of a travel domain may also overlap with a business domain .

Hereinafter, a configuration of a screen for selecting a domain displayed by the display unit 100 according to the present embodiment will be described with reference to the drawings.

In this embodiment, the domain display unit 100 displays the domain that the user can select or deselect. Referring to FIG. 4, the display unit 100 may display domains in a hierarchical or tree structure. As shown in FIG. 4, each domain can be represented by a label which is a name of a domain designated by the domain.

(47), a business domain, a conference domain (45), and a medical domain (46) in four subdomains of a general domain. have.

The travel related domain may be represented by a domain 47 having a label labeled 'Touring' and may include a restaurant domain, an airport domain, and a car rent domain in three subdomains . Further, the restaurant domain may be represented by a domain having a label named 'Restaurant' and may include additional subdomains depending on the type of the restaurant. For example, in FIG. 4, the Korean restaurant domain is labeled 'Korean Food' and the Chinese restaurant domain is labeled 'Chinese Food'.

The Korean restaurant domain may include language data for Korean food and restaurant names to support speech recognition.

Also, the conference domain may be represented by a label 45 named 'Conference' and may include computer engineering domains and mechanical engineering domains as subdomains. Computer engineering domains can be represented by the label 'Computer Science', and the mechanical engineering domain can be represented by the label 'Mechanical Engineering'. The conference domain has a higher frequency of use of specialized vocabulary than other areas, so that it is possible to improve the accuracy of recognition and the accuracy of interpreting when the designated speech recognition service is provided by segmenting the speech recognition area according to the related field of the conference .

Hereinafter, a user input unit 200 receiving a selection of a domain for selecting a domain from a user through the domain display unit 100 will be described.

4, in the present embodiment, the display unit 100 of the user terminal 10 creates a tree structure of domain regions and displays the domain regions to the user, and the user selects a domain 45 to be selected as a designated domain display region The user can select the domain by dragging and dropping 43 to the mouse or touch gesture. In addition, the designation can be canceled by dragging and dropping the domain 44 already selected in the designated domain display area 42 out of the area.

At this time, the domain tree can display or hide the subdomains by the '+' button 46 or the '-' button 47.

Also, the already selected domains among the nodes of the tree may be represented differently so that the user can avoid unnecessary reselection. In addition, since the general area of the selected designated domain display part 42 is selected in advance, it is displayed differently from other domains so that the user can recognize this fact.

4, 'General', 'Touring', 'Restaurant', and 'Korean Food' are displayed as a domain for a designated domain already selected in the assignable area display area of the display unit ) And informs the user that it has been selected. In addition, the selected domain is displayed in the designated domain display area 42. The 'General' domain 48 is basically a selected domain, and can not be deselected. .

4 also shows that the current user drags (43) the 'Conference' domain 45 to the designated domain display area 42 to specify the meeting related domain. Furthermore, in this embodiment, in addition to dragging and dropping the label of the domain for the selection of the designated domain, it is also possible to select the designated domain by a mouse-click or right-click menu call.

Furthermore, referring to FIG. 5, instead of the drag-and-drop method, the user can click on the touch or touch to check that the empty check box 51 is selected. Also, in the case of the general domain (53) which should always be selected, it is possible to notify the user that it is always selected by displaying a check with a different color.

6 shows an example of dynamically representing a user interface for selecting a domain of the present invention. Figure 6 shows the tree structure of Figures 4-5 more dynamically so that the user can manipulate it more easily. The user clicks or touches the desired domain to display the child nodes as subdomains of the corresponding domain, and selection 61 is possible. Also, the domains of the unselected designated domains are represented (62) differently to the user.

7 shows an example of a user interface for specifying designated domains having an equivalent structure without a hierarchical structure. Each designated domain is indicated by an icon. At the bottom of the icon, the label name of the domain is displayed, and the user is informed to which domain the corresponding icon corresponds. In addition, in the case of the icon form, all of the icons are displayed in the same form in FIG. 7, but it is also possible to configure the 'Medical' icon as an intuitive form such as '+' so as to inform the user of the intuitively corresponding domain.

The screen includes a designatable domain display area 74 to inform the user of the selectable domain area. The selected domain display area 71 is displayed at the bottom of the screen, Tell me what it is. The user selects a designated domain by clicking or touching the domain 72 corresponding to the designated domain to be selected in the assignable region display region or by dragging and dropping the selected domain into the selected domain display region 71. The selected designated domain makes the icon disappear in the existing assignable area display area, thereby avoiding unnecessary reselection of the user. Likewise, the already selected selected domain is clicked or touched, or dragged and dropped to the designated domain display area to deselect it. The released designated domain is displayed in the assignable domain display area 74 so that it can be selected again. In addition, the scroll bar 75 is arranged to increase accessibility so that a user can easily access many designated domains. In the present embodiment, the icons displayed in the assignable domain display area are changed by the scroll operation, but in the case of the designated domain display area, it is preferable that they are not changed regardless of the scroll for selection through the icon.

FIG. 8 is an example of a user interface for designated domains having a hierarchical structure using FIG. When the user selects or touches the icon as the domain 81 for the designated domain to which the user wants to view the sub-designated domain, the icons for the sub-domain of the corresponding domain appear in the sub-domain display portion. The lower region display area 82 shows boundaries so as to be distinguished from upper designated domain domains. The icons 84 may also be dragged and dropped into the designated domain display area 83, or selected by using a click or a touch operation. If the upper designated domain is selected / released in the deselection of the designated domain, the lower designated domain automatically included in the higher designated domain is also selected / released to support comprehensive selection / cancellation.

In FIG. 8, when the user selects the icon 81 corresponding to the 'Seoul' domain, the subdomains of the 'Seoul' domain, for example, 'Seoul Hotel', 'Seoul Restaurant' and the like are displayed. The user can select the Seoul hotel related domain as the designated domain by touching the 'Seoul Hotel' 84 among the displayed icons or by dragging and dropping it to the designated domain display area 83.

Further, the display unit 100 according to the present embodiment can identify the user's situation by using the user information collected through the user's terminal, and display the recommended domain to the user according to the poured status information. FIG. 9 shows an example of identifying a user's situation and presenting an appropriate designated domain.

The user information collected through the user's terminal is information about the user's location through a GPS (Global Positioning System) built in the user terminal, peripheral information through the camera, surrounding sound information recognized through a microphone, . Therefore, the speech recognition apparatus according to the present embodiment recommends a domain that can be designated through the information on the user situation to the user. For example, it is possible to recommend Korea's Seoul as a designated domain through the user's GPS. If the user's peripherals are recognized as a restaurant through the camera, it is also possible to recommend a travel and restaurant as a designated domain. You can also recommend an airport if you are aware of the sound of airplanes taking off and landing as a sound through the microphone.

Therefore, in the present embodiment, it is preferable that the display unit 100 emphasizes the domains of the recommended domains to be displayed on the screen. The user can easily select only the required designated domain from among the designated domains, thereby preventing unnecessary addition of recognition support data supporting recognition of vocabulary and sentence expressions, thereby enabling faster and more accurate voice recognition and automatic interpretation result. On the contrary, it is also possible to facilitate recognition of the user and to prevent unnecessary selection by not displaying or displaying the designated domains that are unnecessary or unlikely to be used by using the situation information.

Referring to FIG. 10, in a case where a user's location is recognized as a Yeosu in Korea through current GPS information, an icon 102 (Yeosu Hotel, Yeosu Restaurant, Yeosu Expo) corresponding to a domain related to Yeosu is highlighted and displayed And the icon 103 (Medical) corresponding to the domain having a low degree of association is blurred.

Further, the domain display unit 100 according to the present embodiment may display at least one example recognition data to the user to illustrate to the user the speech recognition level designated according to the selection of the domain.

Referring to FIG. 11, when a user intends to add a 'Conference' related domain, when the 'Conference' domain is selected as a designated domain through the recognition example part 114, the level of recognizable voice is' Where is the nearest Gal- bi buffet from the 8 th Advanced Computing (115) Conference hall? "Can be indirectly illustrated to help users select a specific domain.

The user input unit 200 receives the domain selected from the user through the domain selection screen displayed on the display unit 100 and the communication unit 300 transmits information on the selected domain to the voice recognition server 20 send.

The speech recognition server 20 performs speech recognition by referring to data corresponding to a domain selected by the user, among the reference data for speech recognition stored in the DB 30 through information on the received selected domain. Then, the voice recognition result is transmitted to the user terminal 10.

Although the speech recognition server 20 according to the present embodiment has been described as a system for communicating with the user terminal 10 for speech recognition and receiving results from the speech recognition server, In this case, the communication unit 300 of the user terminal 10 transmits the information on the selected domain to the voice recognition server 20 through the voice recognition module in the terminal, and the DB 30 can be implemented through the internal memory. To the internal speech recognition module rather than to the external speech recognition server 20.

That is, in this case, the configuration of the user terminal 10, the display unit 100, the user input unit 200, and the communication unit 300 operate as an interface module for selecting a speech recognition domain, As shown in FIG.

The user can easily understand and easily modify the designated domain through the speech recognition device 10 according to the embodiment of the present invention, thereby improving the adaptability to the changed environment, And the accuracy of the automatic interpretation can be increased. Hereinafter, an area designation method using the speech recognition apparatus 10 according to the present embodiment will be described.

Referring to FIG. 12, the speech recognition area designation method includes a domain selection screen display step S100, a domain selection input step S200, and a selection information transmission step S300.

The domain selection screen display step S100 displays the screen for selecting the domain as the unit for the speech recognition area of the predetermined classification for the specified speech recognition of the speech recognition server 20 by the display unit 100 do.

In the domain selection input step (S200), the user input unit 200 receives a selection of a domain from a user.

In the selection information transmission step S300, the communication unit 300 transmits the user's selection information for the domain to the voice recognition server 20. [

The detailed operations of each step of the area designation method are the same as those described in the display unit 100, the user input unit 200, and the communication unit 300, and the description thereof will be omitted.

Although the speech recognition operation is described above as an example, the speech recognition operation is also necessary for the automatic interpretation, so that the same can be applied to the automatic interpretation. For example, the speech recognition server 10 of FIG. 1 may be an automatic interpretation server.

The various aspects, embodiments, implementations, or features of the invention described above may be used individually or in any combination, and the various embodiments described herein may be implemented as software, hardware, or a combination thereof, for example have. According to a hardware implementation, the embodiments described herein may be implemented as application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays May be implemented using at least one of a processor, controllers, micro-controllers, microprocessors, and an electrical unit for performing a function. Readable medium may be any data storage device capable of storing data that can be later read by a computer system. Examples of computer readable media include read-only memory, random access memory Memory, CD-ROM, DVD, magnetic tape, optical data storage. Readable medium can also, be stored and executed as computer-readable code is distributed manner, network-may be distributed across computer systems connected.

The present invention has been described with reference to the preferred embodiments. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

Claims (18)

A display unit including a screen displaying a plurality of domains selectable by the user for speech recognition;
A user input unit for receiving a selection of at least one of the plurality of domains from the user; And
And a communication unit for transmitting the selection information of the user to the domain,
Wherein the plurality of domains include at least one other domain including a general domain including a database usable for speech recognition and another database usable for speech recognition. The method according to claim 1,
Wherein the display unit displays a domain selectable by the user and a domain selectable by the user. The method according to claim 1,
Wherein the display unit displays the domain indicating the domain by classifying the domain into a hierarchy according to the speech recognition level. The method of claim 3,
Wherein the display unit displays a domain for the domain selected from the user among the displayed domains classified into hierarchies. The method of claim 3,
Wherein the hierarchical layer according to the speech recognition level is classified according to a generation status of a voice, and the generation status is reclassified according to a generation place. The method of claim 3,
Wherein the display unit displays a domain indicating a domain corresponding to a lower layer of the selected domain according to the selection of the domain of the user. The method according to claim 1,
Wherein the display unit recognizes a user's situation using user information collected through the user's terminal, and displays a recommended domain to the user according to the poured situation information. The method according to claim 1,
Wherein the display unit displays at least one example recognition data to the user to illustrate to the user a speech recognition level designated according to the selection of the domain. A domain selection screen display step of displaying a screen for selecting at least one of a plurality of domains as a unit for a speech recognition area of a predetermined classification for speech recognition;
A domain selection input step of receiving a selection of at least one of the plurality of domains from the user; And
And a selection information transmitting step of transmitting the selection information of the user to at least one of the plurality of domains,
Wherein the plurality of domains include at least one other domain including a general domain including a database usable for speech recognition and another database usable for speech recognition. 10. The method of claim 9,
Wherein the domain selection screen display step displays the domain selectable by the user or the domain selectable by the user. 10. The method of claim 9,
Wherein the domain selection screen display step displays the domain indicating the domain by classifying the domain into hierarchies according to the speech recognition level. 12. The method of claim 11,
Wherein the domain selection screen display step displays a domain of the domain selected from the user among the displayed domains classified into hierarchies. 12. The method of claim 11,
Wherein the hierarchy according to the speech recognition level is a hierarchical structure for classifying a general area providing a basic speech recognition area according to a generation situation of a voice and reclassifying the generation situation according to a generation place. 12. The method of claim 11,
Wherein the domain selection screen display step displays a domain indicating a domain corresponding to a lower layer of the selected domain according to the selection of the domain of the user. 12. The method of claim 11,
Wherein the domain selection screen display step identifies a user's situation using user information collected through the user's terminal and displays a recommended domain to the user according to the spoken context information. 12. The method of claim 11,
Wherein the domain selection screen display step displays at least one example recognition data to the user to illustrate to the user a speech recognition level designated according to the selection of the domain. delete An interface module for displaying a screen for selecting at least one domain among a plurality of domains for speech recognition to the user, receiving a selection of the domain from the user, and transmitting the selection information of the user to the domain; And
And a speech recognition module for performing speech recognition by referring to data corresponding to a domain selected by the user among the reference data for speech recognition through the received selection information of the user,
Wherein the plurality of domains include at least one other domain including a general domain including a database usable for speech recognition and another database usable for speech recognition.

Images