CN101206675B - Messaging device and information processing method - Google Patents

Abstract

The present invention provides a kind of messaging device and information processing method, and this messaging device includes: memory element, is used for storing dictionary data, in this dictionary data, and dependently of each other registration track and one or more content；Input block, the track being associated with content for input；Comparing unit, for comparing the track inputted by described input block and the track being registered in described dictionary data；And adding device, for the comparative result generated based on described comparing unit, add data to the described dictionary data being stored in described memory element.

Description

Messaging device and information processing method

Technical field

The present invention relates to one to use and comprise content and be associated track (locus) The messaging device that content is scanned for by dictionary data (dictionary data).More By the track identified, content can be searched in particular it relates to a kind of The messaging device of rope.

Background technology

Traditionally, content can be carried out point according to the metadata being previously attached in content Class.Japanese Unexamined Patent Publication 2005-236646 publication describes a kind of spy according to thumbnail The amount of levying carries out the image display classified, wherein, under the conditions of Characteristic Extraction from The thumbnail data of view data extracts the characteristic quantity of thumbnail.

Additionally, Japanese Unexamined Patent Publication 10-171835 publication describes a kind of method, its In, messaging device uses search key to scan for, and this search key is Such as date created and time, input coordinate position is used by search key signal generating unit Put, show that area, live width and writing speed generate.When user operation pen also This data are obtained during input stroke data (stroke data).Also can be from the knot of this operation Fruit obtains this data.

Additionally, Japanese Unexamined Patent Publication 07-073190 publication describes a kind of pen control type (pen-based) computer system, wherein, uses the picto-diagram picture being used as filename (pictogram image) scans for.

But, according to figure described in Japanese Unexamined Patent Publication 2005-236646 publication As display device, when characteristic quantity is similar, classification becomes incorrect.It addition, many In the case of number, when classifying based on image, it is impossible to produce as user is desired Raw classification results.

Additionally, at information described in Japanese Unexamined Patent Publication 10-171835 publication Reason equipment needs user to store the operation of input coordinate position or date created and time etc. Process.

Additionally, pen control type described in Japanese Unexamined Patent Publication 07-073190 publication Computer system is used as the picto-diagram picture of filename owing to search for from All Files, So that certain process time.

It addition, in the said equipment and system, even if when inner capacities increases, classification Method still keeps constant.

Summary of the invention

Embodiments of the invention relate to a kind of easily carry out reflecting user view simple point The messaging device of class.

According to an aspect of the present invention, it is provided that a kind of messaging device, this information Processing equipment includes: memory element, is used for storing dictionary data, at described dictionary number According to, dependently of each other registration track and one or more content；Input block, uses In the track that input is associated with content；Comparing unit, is used for comparing by described input The track of unit input and the track being registered in described dictionary data；And add single Unit, for based on the comparative result generated by described comparing unit, described to being stored in Described dictionary data in memory element adds data.

According to a further aspect in the invention, it is provided that a kind of equipment, this equipment includes: deposit Storage unit, is used for storing dictionary data, in described dictionary data, dependently of each other Registration voice and one or more content；Input block is relevant to content for input The voice of connection；Comparing unit, for compare the voice that inputted by described input block and It is registered in the voice in described dictionary data；And adding device, for based on described The comparative result that comparing unit generates, to the institute's predicate being stored in described memory element Allusion quotation data add data.

According to another aspect of the invention, it is provided that a kind of method, the method includes: deposit Storage dictionary data, in described dictionary data, dependently of each other registration track and Or a plurality of content；The track that input is associated with content；Relatively the track of input with step on Note track in described dictionary data；And based on comparative result to described dictionary number According to adding data.

According to an exemplary embodiment of the present invention, due to afterwards can be without key entries Or in the case of menu setecting, add class or type, therefore can realize reflection simply and use The classification that family is intended to.Further, since utilize dictionary number by the input trajectory identified Scan for according to the content classified, therefore scan for phase with to all the elements Ratio, can more quickly perform this search.

By below with reference to the accompanying drawing detailed description to exemplary embodiments, its of the present invention Its feature and aspect will become clear from.

Accompanying drawing explanation

The accompanying drawing of the part comprising in the description and constituting this specification illustrates this Bright exemplary embodiments, feature and aspect, and it is used for explaining this together with description The principle of invention.

Fig. 1 shows the messaging device according to the present invention the first exemplary embodiments Use form；

Fig. 2 shows the messaging device according to the present invention the first exemplary embodiments Block diagram；

Fig. 3 shows the messaging device according to the present invention the first exemplary embodiments The figure of structure；

Fig. 4 shows by the messaging device according to the present invention the first exemplary embodiments The example of shown content choice picture；

Fig. 5 shows by the messaging device according to the present invention the first exemplary embodiments The example of the full screen display of shown selected content；

Fig. 6 shows the track classified dictionary according to the present invention the first exemplary embodiments The figure of the data structure of (locus classification dictionary)；

Fig. 7 shows the track to content according to the present invention the first exemplary embodiments The flow chart of the example of input processing；

Fig. 8 shows by the messaging device according to the present invention the first exemplary embodiments The example of track input picture during shown search；

Fig. 9 shows by the messaging device according to the present invention the first exemplary embodiments The example of shown Search Results display picture；

Figure 10 shows searching of the use track according to the present invention the first exemplary embodiments The flow chart that rope processes；

Figure 11 shows the messaging device according to the present invention the second exemplary embodiments The figure of structure；

Figure 12 shows the shape of the input trajectory according to the present invention the second exemplary embodiments Example with the shape of dictionary track；

Figure 13 shows showing of the dictionary data according to the present invention the second exemplary embodiments Example；

Figure 14 shows the additional data that has according to the present invention the second exemplary embodiments The example of dictionary data；

Figure 15 shows the track to content according to the present invention the second exemplary embodiments The flow chart of the example of input processing；

Figure 16 shows the messaging device according to the present invention the 3rd exemplary embodiments The figure of structure；

Figure 17 shows the Classification of Speech dictionary according to the present invention the 3rd exemplary embodiments The figure of the data structure of (speech classification dictionary)；

Figure 18 shows the voice to content according to the present invention the 3rd exemplary embodiments The flow chart of the example of classification Information input processing；

Figure 19 shows in the use voice according to the present invention the 3rd exemplary embodiments Hold the flow chart of the example that search processes.

Detailed description of the invention

The various exemplary embodiments of the present invention, feature is described in detail below with reference to accompanying drawing And aspect.

First shows exemplary embodiments

Fig. 1 shows the messaging device 104 according to the present invention the first exemplary embodiments Use form.DTV 101 includes liquid crystal display, CPU (CPU), read only memory (ROM), random access memory (RAM), hard drive Device (HDD), record medium slot and for be connected to can be connected to the network The NIU of people's computer.Additionally, DTV 101 includes in difference Reproduce on network and show content-data built-in including rest image or moving image Software.

LAN 102 be used for being connected to hdd recorder etc. for storing static image or The storage device of moving image.By LAN 102 by media server (media Server) 103 it is connected to messaging device 104.Rest image or moving image are deposited Storage is in media server 103.According to the request sent by LAN 102, media Server 103 transmits the data of such as rest image etc. by LAN 102.

Messaging device 104 includes liquid crystal display, arranges on a liquid crystal display The number of such as transparent resistance digitizer (transparent resistive digitizer) etc. Word instrument, CPU, memorizer, WLAN (LAN) chip etc..Pen 105 is permissible As digitizer.User can use pen 105 to display on liquid crystal display Rest image adds track.

Fig. 2 shows the block diagram of the messaging device 104 according to this exemplary embodiments. Position coordinates input block 201 includes transparent resistance digitizer.Such as, when pressing with pen During the surface of lower position coordinates input block 201, will be pressed by system bus 207 The XY coordinate data of position is sent to CPU 203.

Liquid crystal display 202 includes liquid crystal display cells, liquid crystal control circuit and shows Show memorizer.By system bus 207, liquid crystal display 202 is connected to CPU 203.Liquid crystal display 202 can show figure according to the instruction from CPU 203 Picture.CPU 203 is connected to RAM 204, ROM 205, liquid by system bus 207 Brilliant display unit 202 etc..CPU 203 holds according to the program being stored in ROM 205 Row processes.The process of flow charts described below is performed by CPU 203.

RAM 204 is used as working area.ROM 205 stores the rail for classifying content The process that mark generates.Flash memory (flash memory) 206 can serve as public information equipment. Owing to flash memory 206 is connected to CPU 203 by system bus 207, thus will include quiet Only image or the local content file of moving image and configuration information are stored in flash memory 206 In.

System bus 207 for CPU 203, ROM 205, RAM 204 and its Send between its device and receive data.WLAN (LAN) chip 208 makes it possible to Enough communicate from different personal computers.Although it should be noted that, implementing typical case In example, use rest image is as the example of content, it is also possible to use moving image, Sound, text, thumbnail or file are as content.

Fig. 3 shows the structure of the messaging device 104 according to this exemplary embodiments Figure.Position coordinates input block 301 includes transparent resistance digitizer.Position coordinates Input block 301 detects the position coordinate data of the track inputted by pen and by this position The CPU 203 that coordinate data is sent in Fig. 2.CPU 203 is to this position coordinate data Carry out the given zone of the RAM 204 in fig. 2 of the data storage after processing and processing In territory.Track identification processing unit 303 will send from position coordinates input block 301 The track characteristic registered in the shape facility of track data and track classified dictionary 304 Data compare, and export the candidate's dictionary data corresponding to this comparative result (candidate dictionary data) and similarity.

Characteristic and track that track classified dictionary 304 comprises track write content Fileinfo.Newly register judging unit 305 according to by track identification processing unit 303 institute The recognition result provided, it is judged that newly register input trajectory or to existing dictionary data Interpolation information.Track classification information adding device 306 is to existing track classified dictionary data Add fileinfo and the positional information of input trajectory of Current Content.

New track dictionary signal generating unit 307 using the characteristic of input trajectory as dictionary Data add track classified dictionary 304 to.Content-data selects unit 308 to select to comprise The file of content also generates and shows this document on the screen of liquid crystal display 302 Instruction.Liquid crystal display 302 shows by selected by content-data selection unit 308 Content, and also display input trajectory.

Fig. 4 shows the example of the content choice picture shown by information process unit.In Hold and select picture to include the viewing area 401 for displaying content list.Scroll bar 402 can For rolling picture.User can operate rolling by position coordinates input block 301 Bar 402 is with rolling picture.In the exemplary screen shown in Fig. 4, content 403 is selected The content selected.If user input select content order, then information process unit with Screen mode toggle show selected by content.

Fig. 5 shows that the full frame of the selected content 501 shown by information process unit shows The example shown.Track 502 is write (picture) to content 501.User can be by Fig. 1 Track the most freely write by pen 105.Track owing to being write is different from content number According to, therefore user can track that freely show or hide is write.

Fig. 6 shows the data of the track classified dictionary 304 according to this exemplary embodiments The figure of structure.The dictionary data of each track comprises track characteristic data 601, multiple track Picture position information and fileinfo 602～604.

Such as, track classified dictionary 304 includes that the track representing the track 605 of "○" is special Levy data, on it with the track input position in the file " 01.jpg " of (being decorated with) "○" Coordinate data and on it with the track input bit in the file " 02.jpg " of "○" The coordinate data put.If only existing a file for specific feature data, the most only List a bar track picture position information and fileinfo.As in track characteristic Shown in the row of " △ ", register on it defeated with the track in the file " 04.jpg " of " △ " Enter the coordinate data of position.Obviously can from the data structure of track classified dictionary 304 Go out, user the track of the arbitrary shape inputted is stored in track classified dictionary 304 In.

Fig. 7 shows the flow chart of the example of the track input processing to content.When Having drawn input trajectory in content and time user asks to add this input trajectory, track inputs Process starts.Such as, add track as user by pen 105 and press carriage return (enter) During key (not shown), this process starts.

In step 701, CPU 203 starts the process to content input trajectory.Initially Change the storage buffer of dictionary data.In a step 702, CPU 203 stores current aobvious The filename of the content shown.Such as, in the case of fig. 5, CPU 203 stores The filename (06.jpg) of the photo of flower.In an embodiment, this document is local file. But, such as, if this document is on network, then CPU 203 stores on network Filename (such as, //net.com//video/06.jpg).

In step 703, the track of the content that CPU 203 is added on picture The serial data of XY coordinate data stores in work buffers.In step 704, CPU 203 characteristics extracting this track from the XY coordinate data string stored.Permissible Tradition track recognizer is used to carry out this extraction.Owing to registering the pen of user Mark and user oneself use this person's handwriting, therefore it is desired that high discrimination.Such as, Input trajectory is standardized as 256*256, track stroke segmentation is become 20 Part, and such as store the position coordinates of each partitioning portion.

In step 705, the character shape data in dictionary data is loaded into by CPU 203 In buffer.Such as, in figure 6, loading includes track "○", " △ ", " ∠ " AndTrack characteristic data 601.In step 706, CPU 203 judges at word Whether allusion quotation data exist the shape similar to the characteristic of input data.If deposited In the dictionary data (being "Yes" in step 706) of similar shape, then process enters step 707.If there is no the dictionary data (being "No" in step 706) of similar shape, Then process enters step 708.By by the characteristic and the dictionary data that input data Characteristic match and extract between characteristic difference to carry out this judgement. If the characteristic of input data is (most like with the dictionary data with minimum difference Dictionary data) characteristic between difference close to " 0 ", then CPU 203 is judged as There is the dictionary data of similar shape.If the characteristic of input data and each dictionary Difference between the characteristic of data is all higher than or equal to given threshold value, then CPU 203 are judged as not existing the dictionary data of similar shape.

In step 707, the characteristic number of the CPU 203 similar shape in dictionary data According to adding track picture position information and fileinfo.Such as, with regard to the track 502 in Fig. 5 For, due to the shape that track 502 is similar "○", therefore track 502 is judged as Match with the shape 605 ("○") in the dictionary data shown in Fig. 6.Therefore, by file The minimum and maximum XY coordinate data of the boundary rectangle of name " 06.jpg " and this input trajectory Add the dictionary data of file " 02.jpg " next column to.

In step 708, CPU 203 display asks the user whether to add to dictionary data The message of characteristic.Screen showing, this message is to prevent user from registering input not Successfully track.Such as, this message is for " to record new track？It is, by " determination ", no Then by " cancellation ".”.Furthermore it is possible to arrange this message by system, with not Show it again.

In step 709, CPU 203 adds character shape data to dictionary data. The CPU 203 track classified dictionary 304 in Fig. 3 adds and is extracted in step 704 Character shape data.Such as, if dictionary data is all dictionary numbers as shown in Figure 6 According to, then this feature data are added to the shape data being listed in dictionary data end Lower section.So, the data added will become the end of dictionary data.

In step 720, CPU 203 is to the dictionary data added in step 709 Add track picture position information and fileinfo.Such as, if by the rail of shape " 3 " Mark adds on the photo " 06.jpg " of the flower in Fig. 5, then by character shape data " 3 " It is registered in character shape dataLower section.Then, registration documents name " 06.jpg " and The maximum XY coordinate data of the boundary rectangle of this input trajectory and minimum XY number of coordinates According to.In step 711, CPU 203 terminates the track input processing to content.

Fig. 8 shows by track input picture during search shown by messaging device The example in face.Fig. 9 shows and is shown picture by the Search Results shown by messaging device The example in face 901.Window 801 in Fig. 8 is the track input window when search.As Fruit presses cancel key 802, then cancel this search and process.If pressing clear key 803, Then delete track shown on track input area 805.If pressing search key 804, then start search based on track shown on track input area 805.When When pressing search key 804, start the process of flow chart in Figure 10.

Figure 10 shows the flow chart of the example using the search of track to process.In step In rapid 1001, CPU 203 begins to use the search of track process and carry out buffer Distribution and initialization.In step 1002, CPU 203 is loaded into the XY of input trajectory Coordinate data.By the track of the track of input in track input area 805 in fig. 8 Data are loaded in buffer.In step 1003, CPU 203 is from loaded rail Mark extracting data character shape data.The XY coordinate figure of track data is standardized as 256*256.Then, track stroke segmentation is become 20 parts, and will be each The position coordinates of partitioning portion stores as serial data.

In step 1004, dictionary data is loaded into work buffers by CPU 203. Such as, by shown in Fig. 6 track characteristic data 601 ("○", " △ ", " ∠ ",) It is loaded into work buffers.

In step 1005, CPU 203 detection is special with input trajectory shape most similar Levy.The XY coordinate of characteristic that CPU 203 will be extracted from input trajectory data Data are compared with the XY coordinate data in dictionary data, and then computed range is poor.Will The dictionary data with lowest difference regards the dictionary number most like with input trajectory data as According to.If input data are the shape inputted in track input area 805 in such as Fig. 8 Shape "○", then track 605 ("○") is judged as most like with these input trajectory data Dictionary data.

In step 1006, the track of detected shape facility is drawn by CPU 203 Face XY coordinate information and fileinfo are loaded in buffer.Such as, shown in Fig. 6 In the case of, the seat of the track input position being loaded in the file " 01.jpg " comprising "○" The number of coordinates of the track input position in the file " 02.jpg " marking data and comprise "○" According to.

In step 1007, CPU 203 calculates input trajectory by equation below The area S of boundary rectangle.

S=dx*dy (xmax xmin=dx, ymax ymmin=dy)

In step 1008, the order that CPU 203 amasss S according to track area approaching face is right Fileinfo in dictionary data is ranked up, and this area S is the external of input trajectory The area of rectangle.Such as, area is divided into three kinds of sizes, the most greatly, neutralizes little.As Really the track area of input trajectory is big, then by fileinfo according to from large area to little The order sequence of area.

Comprise in the case of there is the data item of equal area size in dictionary data, In step 1009, CPU 203 compares the center of the boundary rectangle of input trajectory and has In the boundary rectangle of the track of each data item in the dictionary data of equal area size The heart.Then, CPU 203 is according to being closely located to the order of input trajectory to fileinfo It is ranked up.Such as, if track be the upper left corner of picture input and dictionary number Track according to is positioned at the center of picture, the lower right corner and the upper left corner, then by fileinfo Sort according to the order at the upper left corner, center to the lower right corner.

In step 1010, CPU 203 filename based on the dictionary data after sequence It is loaded into file data.Such as, if the file after Pai Xu is file shown in fig .9 01.jpg, 02.jpg and 06.jpg, then CPU 203 is loaded into the file data of these files.

In step 1011, CPU 203 is based on loaded file generated thumbnail figure Picture also shows this thumbnail image.Such as, if loaded file is file 01.jpg, 02.jpg and 06.jpg, then according to these file generated thumbnail images, and With thumbnail 902 (01.jpg) as shown in Figure 9, thumbnail 903 (02.jpg) and thumbnail The order display thumbnail image of 904 (06.jpg).Track is together with each thumbnail image Display.In step 1012, process terminates.

According to above-mentioned process, messaging device can use has arbitrary shape The content-data of such as rest image etc. is classified by any type of track.With this The mode of kind, owing to can add afterwards in the case of without key entries or menu setecting Add class or type, therefore can realize reflecting the classification of user view simply.It addition, The input trajectory identified owing to utilizing dictionary data the to pass through content to being classified is carried out Search, more quickly performs this search when therefore can scan for comparison all the elements.

Second exemplary embodiments

According to the first exemplary embodiments, input trajectory data and dictionary data have one Threshold value is as criterion.Therefore, user must select alternatively is to newly register Or add data to existing track dictionary.But, according to the person's handwriting of user, The trajectory shape registered in dictionary data is probably unclear.In this case, Input trajectory data can not be well matched with dictionary data.In order to prevent this problem, According to exemplary embodiments, determine that two threshold values are as criterion.If matching degree is little In threshold value " 1 ", then add new data to existing track dictionary data.If coupling The data of similar shape equal to " 1 " or more than " 1 " but less than " 2 ", are then added to by degree Existing shape data.Use form due to the messaging device of this exemplary embodiments With block diagram be similar to the first exemplary embodiments shown in Fig. 1 and Fig. 2 use form and Block diagram, thus will not be repeated again the explanation to these contents.

Figure 11 shows the messaging device according to the present invention the second exemplary embodiments The structure of 104.Due to unit 1101～1108 functionally with the first allusion quotation shown in Fig. 3 The unit 301～308 of type embodiment is similar to, and therefore will not be repeated again these contents Explanation.Similar shape information adding device 1109 is used for adding shape and existing shape phase As different pieces of information.

Figure 12 shows the shape of the track in the shape of input trajectory and dictionary data Example.Track 1201 and 1204 is the example of input trajectory data.Track 1202 and 1203 Example for the track data in dictionary data.

Figure 13 shows the example of dictionary data.Figure 14 shows that having one is added The example of dictionary data to the trajectory shape data of the dictionary data shown in Figure 13.

Figure 15 shows the example of the track input processing according to this exemplary embodiments Flow chart.

Process step 1501～the process of 1505,1511 and 1512～1514 in Figure 15 divide Not with the step 701 in Fig. 7 of the first exemplary embodiments～705,707 and 708～710 classes Seemingly.Therefore, explanation to these steps is not repeated.

In step 1506, CPU 203 detects characteristic closest to input trajectory The dictionary data of characteristic.Such as, CPU 203 compares the " dictionary data in Figure 12 1 " 1202 and " input data 1 " 1201.CPU 203 is by by the track after standardization The range difference of individual features point is added together to calculate " dictionary data 1 " 1202 and " input Data 1 " diversity factor between 1201.Then, CPU 203 stores the difference calculated Different degree.Then, CPU 203 compares " dictionary data 2 " 1203 and " input data 1 " 1201. Then, CPU 203 calculates and stores " dictionary data 2 " 1203 and " input data 1 " 1201 Between diversity factor.Additionally, CPU 203 compares the diversity factor of " dictionary data 1 " 1202 The diversity factor of " dictionary data 2 " 1203 also determines the characteristic number closest to input trajectory According to dictionary data.The dictionary data that diversity factor is less is the spy closest to input trajectory Levy the dictionary data of data.

In step 1507, CPU 203 determines the characteristic closest to input trajectory Diversity factor DST of dictionary data.

In step 1508, CPU 203 will diversity factor determined by step 1507 DST compares with threshold value " 1 " (MiniDst1).If diversity factor DST is less than threshold value " 1 " (being "Yes" in step 1508), then process enters step 1511.If diversity factor DST equals to or more than threshold value " 1 " (being "No" in step 1508), then process enters step Rapid 1509.If the characteristic of the characteristic of input trajectory and dictionary data as " input data 1 " 1201 in Figure 12 are the most similar with " dictionary data 1 " 1202, the poorest Different degree DST is less than threshold value " 1 " and the process being suitable in step 1511.

In step 1509, CPU 203 will diversity factor determined by step 1507 DST compares with threshold value " 2 " (MiniDst2).If diversity factor DST is less than threshold value " 2 " (being "Yes" in step 1509), then process enters step 1510.If diversity factor DST equals to or more than threshold value " 2 " (being "No" in step 1509), then process enters step Rapid 1512.Such as, " the input data 2 " difference between 1204 and " dictionary data 1 " 1202 Different degree is more than " input data 1 " diversity factor between 1201 and " dictionary data 1 " 1202. In this case, diversity factor DST is equal to or more than threshold value " 1 " but less than threshold value " 2 ". Therefore, the process being suitable in step 1510.

In step 1510, CPU 203 adds input shape to the dictionary data mated Shape characteristic.With reference to Figure 12～Figure 14, this process will be described.As shown in Figure 13, Add input character shape data before, dictionary data include a shape data, The coordinate data of the track input position in the file " 01.jpg " comprising " 3 " and comprising The coordinate data of the track input position in the file " 02.jpg " of " 3 ".

Using the track of " the input data 2 " 1204 in Figure 12 as input data input. " input data 2 " diversity factor between 1204 and " dictionary data 1 " 1202 equals to or more than Threshold value " 1 " but less than threshold value " 2 ".Then, by the input shape of " input data 2 " 1204 Characteristic is added to after the shape data of " dictionary data 1 " 1202.Figure 14 illustrates There is the dictionary data of additional data.Dictionary data shown in Figure 14 includes " dictionary Data 1 " 1202 character shape data 1401, from current " the input data added 2 " 1204 character shape datas 1402 generated, comprise in the file " 01.jpg " of " 3 " The coordinate data 1403 of track input position, the rail comprised in the file " 02.jpg " of " 3 " The coordinate data 1404 of mark input position and comprise " 3 " and the file added for this The coordinate data 1405 of the track input position in " 03.jpg ".

As it has been described above, when the shape instability of the track being registered in dictionary, permissible The efficiency of classification and search is improved by accumulation shape data.

3rd exemplary embodiments

According to the first and second exemplary embodiments, messaging device is for by input Content is classified and searches for by the track of arbitrary shape.Such as, existed by user Content is watched, it is also possible to be used for by the input block realizing voice application point on television set Class and search content.In this case, with voice-input unit and Classification of Speech word Allusion quotation replaces position coordinates input block 201 and the track classificating word of the first exemplary embodiments Allusion quotation 304.By replacing position coordinates input block 201 with voice-input unit, can Use form in Fig. 1 and the block diagram in Fig. 2 are applied to according to this exemplary embodiments Messaging device.

Figure 16 shows the structure of the messaging device according to this exemplary embodiments Figure.Voice-input unit 1602 includes mike and analog to digital (AD) transducer.Will The speech capturing of input is numerical data.Voice match processing unit 1603 performs input Mating between feature and the feature of dictionary data of speech data.Can will include tradition The suitable matching algorithm of speech recognition algorithm is used for this process.

Classification of Speech dictionary 1604 includes the voice feature data of registration and at registration voice The filename of the content-data obtained during feature.If at Classification of Speech dictionary 1604 In register the speech data corresponding with inputting speech data, then newly register judging unit 1605 call Classification of Speech information adding device 1606.If at Classification of Speech dictionary In 1604, the unregistered speech data corresponding with input speech data, then newly register judgement Unit 1605 calls new speech dictionary signal generating unit 1607.

Classification of Speech information adding device 1606 is worked as adding to the speech data of coupling Front content information.New speech dictionary signal generating unit 1607 is for Classification of Speech dictionary 1604 features adding newly inputted speech data.Content-data selects unit 1608 to select Select and generate the instruction of displaying content file on the screen of liquid crystal display 1601.

Figure 17 shows the data structure of Classification of Speech dictionary 1604.One phonetic dictionary Data include voice feature data 1701, speech volume when representing phonetic entry and each The speech volume information of the filename of content and fileinfo 1702～1704.Such as, If input voice " is spent " when the photo of display flower, then record the phonetic feature of " spending " With title " 13.jpg (comprising the filename of the file of colored photo) ".

Figure 18 shows the stream of the example of the Classification of Speech Information input processing to content Cheng Tu.Before reason starts in this place, user shows this content, picks up mike, presses Lower start button, to microphone talk, such as " Mike ", and press record button. Then, speech data it is digitized and starts to process.In step 1801, CPU 203 start the phonetic entry to content processes.CPU 203 distributes depositing of storage buffer Storage area.In step 1802, CPU 203 stores the filename of currently displaying content. If the file " 13.jpg " of the currently displaying photo as flower, then CPU 203 stores literary composition Part name " 13.jpg ".

In step 1803, input speech data is loaded in buffer by CPU 203. In step 1804, CPU 203 extracts characteristic number from loaded input speech data According to.Such as, as described in Japanese Unexamined Patent Publication 2005-175839 publication, will Speech data is converted to characteristic parameter string.In step 1805, CPU 203 is by Figure 17 Shown in dictionary data be loaded in buffer.

In step 1806, CPU 203 judges whether deposit in dictionary data and input language The speech data that sound data are similar.If being found that similar voice in dictionary data Data (being "Yes" in step 1806), then process enters step 1807.If at word Allusion quotation data do not find similar speech data (for "No" in step 1806), then locates Reason enters step 1808.CPU 203 carries out the characteristic parameter string of this input speech data And mating between the characteristic parameter string of the speech data registered.If having the highest The similarity of the dictionary data of similarity is more than predetermined threshold value, then CPU 203 judges Similar to this dictionary data for this input speech data.

In step 1807, CPU 203 adds voice to the phonetic dictionary data of coupling Information volume and content file information.Such as, if when watching file " 16.jpg ", User says " Mike " facing to mike with the audio volume level of 25 decibels (dB), then this input Speech data and the phase of the voice feature data " Mike " in the dictionary data in Figure 17 Join.Therefore, the information that is close in the dictionary data in Figure 17 " speech volume 30dB, 12.jpg " add information " speech volume 25dB, 16.jpg ".

In step 1808, CPU 203 shows on screen and is used for adding new speech number According to message.Such as, this message is for " to add this voice？It is, by " determination ", otherwise By " cancellation ".”.If selected for determining, then process enters step 1809.If choosing Selected cancellation, then process enters step 1811 and non-registration new speech data.

In step 1809, CPU 203 adds input speech data to dictionary data Characteristic parameter.If when dictionary is in the state shown in Figure 17, user's input is " big Sea ", then the voice feature data being not present in the dictionary of Figure 17 due to " sea " In, therefore voice feature data " sea " is added to the dictionary data " wind in dictionary Scape " after.

In step 1810, CPU 203 adds voice sound to this new speech characteristic Amount information and content file information.Such as, if the user while watch the photo in sea Time " 16.jpg ", with 30dB input speech data " sea ", then due to the feature in sea It is not present in the phonetic feature dictionary of Figure 17, therefore, voice feature data is " big Sea " add the dictionary data " landscape " in dictionary to after.Additionally, also by information " language Sound volume 30dB, 16.jpg " add this new speech characteristic to.According to above-mentioned process Generate Classification of Speech dictionary 1604.

Figure 19 shows the flow process using voice that content scans for the example processed Figure.When user inputs to be searched for after have selected " phonetic entry " from searching menu During the voice of content, start content search and process.

In step 1901, CPU 203 uses voice to start search and processes.CPU 203 Carry out the distribution of storage buffer.In step 1902, CPU 203 will act as search The input speech data of the search key processed is loaded in buffer.In step In 1903, CPU 203 extracts characteristic from the speech data being loaded into.Such as, As described in Japanese Unexamined Patent Publication 2005-175839 publication, CPU 203 is by language Sound data are converted to characteristic parameter string.

In step 1904, CPU 203 determines the audio volume level of input speech data. Such as, if user produces voice with 20dB, then CPU 203 determines that audio volume level is 20dB.In step 1905, dictionary data is loaded in buffer by CPU 203. Such as, the dictionary data shown in Figure 17 is loaded in buffer.In step 1906 In, CPU 203 is according to the voice feature data in dictionary data and input phonetic feature Dictionary data is sorted by the similarity between data.If user inputs such as " landscape " Deng voice and scan for, then data " landscape " and " speech volume 20dB, 15.jpg " Will appear in the top of the dictionary data shown in Figure 17.

In step 1907, CPU 203 is according to the volume water being included in dictionary data Dictionary data is sorted by ordinary mail breath with the comparison of input speech volume information.Such as, logical Cross the phonetic entry " Mike " of 30dB, by dictionary data according to data " speech volume 30 DB, 12.jpg " and the order sequence of " speech volume 20dB, 11.jpg ".

In step 1908, CPU 203 generate and show with sequence after dictionary data The thumbnail image of corresponding content.Such as, if the result of search is data " voice Volume 30dB, 12.jpg " and " speech volume 20dB, 11.jpg ", then CPU 203 reads Take the view data of file " 12.jpg ", generate its thumbnail image on a display screen Show this thumbnail image.Then, CPU 203 reads the picture number of file " 11.jpg " According to, generate its thumbnail image and show this thumbnail image on a display screen.? In step 1909, CPU 203 terminates to be processed by the search of voice.

Utilize said structure, messaging device to may utilize voice to be carried out by content-data Classification also scans for effectively.Owing to the voice that user produces is registered as voice number According to, therefore can realize higher search efficiency.

Can also be come real by such as system, equipment, method, program or storage medium Existing above-mentioned exemplary embodiments.Additionally, also by the system or the bag that include multiple device Include the equipment of single assembly to realize above-mentioned exemplary embodiments.

The present invention includes situations below: when directly or being carried from remote site by software program Feed system or equipment, and the journey that the computer execution of this system or equipment is provided When sequence code is to realize the function of above-mentioned exemplary embodiments, it is achieved above-mentioned exemplary embodiments Function.In this case, the program provided corresponds to above-mentioned exemplary embodiments In figure shown in flow chart.

Therefore, program code is being installed in computer and by computer to perform it In the case of realizing the function of the present invention, program code itself also constitutes the present invention. In other words, the present invention includes the process of the function for realizing above-mentioned exemplary embodiments Computer executable program.In this case, if object code, by explain journey Program that sequence performs or be supplied to the script data etc. of operating system and play the work of program With, then can provide computer executable program with these forms.

For providing the record medium of this program to include, such as, floppy disk, hard disk, light Dish, magneto-optic disk (MO), compact disk read only memory (CD-ROM), can imprinting light Dish (CD-R), CD-RW (CD-RW), tape, Nonvolatile memory card, only Read memorizer (ROM), digital versatile disc (DVD) (the read-only storage of digital versatile disc Device (DVD-ROM) and can imprinting digital versatile disc (DVD-R)).

Also can be supplied to this program use client computer browser to be connected to Yin Te Net website and the computer executable program of the present invention is downloaded to the note of such as hard disk etc. The user of recording medium.In this case, the program downloaded can be to have automatically The compressed file of installation function.

Furthermore, it is possible to the program code of the program constituting exemplary embodiments of the present invention is divided Become multiple file, and download each file from different internet websites.In other words, Multiple user is allowed to download the program file of the function for realizing above-mentioned exemplary embodiments WWW (WWW) server also constitute the present invention.

Furthermore, it is possible to by the program encryption of the present invention, record the note at CD-ROM etc. On recording medium and be distributed to user.In this case, it is allowed to meet predetermined condition User downloads encryption key message by the Internet from internet website, to use this to add Key information deciphers encrypted program, and installs on computers and to be deciphered Program.

When performing provided program by computer, it is achieved that above-mentioned typical case implements The function of example.It addition, this program can be together with running operating system on computers Use with the function realizing above-mentioned exemplary embodiments.In this case, operating system Carry out whole actual treatment or part actual treatment.Above-mentioned typical case is realized according to this process The function of embodiment.

Additionally, the program from record medium reading is storable in as being connected to computer To realize above-mentioned typical case in memorizer provisioned in expansion board or functional expansion unit All or part of function of embodiment.In this case, this program is being stored in After in expansion board or functional expansion unit, instruction based on this program, for merit CPU provisioned in energy expansion board or functional expansion unit performs whole actual treatment or portion Divide actual treatment.

Although illustrating the present invention with reference to exemplary embodiments, it is to be understood that, the present invention It is not limited to disclosed exemplary embodiments.The scope of the appended claims meets Wide explanation is to contain all modifications, equivalent structure and function.

Claims (6)

1. a messaging device, it is characterised in that including:

First input block, the first track being associated with content for input；

Memory element, is used for storing dictionary data, in described dictionary data, with interior Hold the first track registering described first input block input explicitly；

Comparing unit, for comparing by the first track of described first input block input With the first track being registered in described dictionary data；

Adding device, for based on the comparative result generated by described comparing unit, to The described dictionary data being stored in described memory element adds data；

Second input block, for input the second track；

Detector unit, for examining from the first track being stored in described dictionary data Survey first rail most like with the shape of the second track of described second input block input Mark；

Sequencing unit, outside the first track detected based on described detector unit Connect the area external square close to the second track of described second input block input of rectangle The order of the area of shape, to relative with the first track that described detector unit detects The multiple contents answered are ranked up；And

Display unit, the content after display is sorted with described sequencing unit is corresponding Data.

Messaging device the most according to claim 1, it is characterised in that as According to the comparative result that described comparing unit is generated, fruit is judged as that described first input is single The first track that unit is inputted is the first rail being previously registered in described dictionary data Mark, the most described adding device adds and the first track pair of input to described dictionary data The information of the content answered.

Messaging device the most according to claim 1, it is characterised in that as According to the comparative result that described comparing unit is generated, fruit is judged as that described first input is single The first track that unit is inputted is not the first rail being previously registered in described dictionary data Mark, the most described adding device adds the first track of input to institute as new track State dictionary data.

Messaging device the most according to claim 1, it is characterised in that institute Predicate allusion quotation data include positional information, size information and the shape phase with the first track The information of the content of association registration.

Messaging device the most according to claim 1, it is characterised in that as Fruit by described first input block input the first track be registered in described dictionary data In the first track between diversity factor equal to or more than the first predetermined threshold but less than the Two predetermined thresholds, the most described adding device be registered in described dictionary data The mode that the shape of one track is associated, additional registration is defeated by described first input block The first track entered.

6. an information processing method, it is characterised in that including:

The first track that input is associated with content；

Storage dictionary data, in described dictionary data, registers explicitly with content First track；

The first track relatively inputted and the first rail being registered in described dictionary data Mark；

Data are added to described dictionary data based on comparative result；

Input the second track；

Second detecting and inputting from the first track being stored in described dictionary data The first track that the shape of track is most like；

The area of boundary rectangle based on the first track detected is close to the second of input The order of the area of the boundary rectangle of track, to relative with the first track detected The multiple contents answered are ranked up；And

Show the data corresponding with the content after sequence.