JP2015127894A - Support apparatus, information processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the following problem: when modification of subtitles is repeated by a plurality of persons, when to conclude the modification cannot be determined properly, or if only the modification is repeated, the modification is trapped in local optimization and the quality of the subtitles cannot be improved sufficiently.SOLUTION: The support apparatus aids in creation of expression information by a plurality of users, comprising: an editing unit for having first expression information expressing an expression target edited by any one of the plurality of users to obtain the edited information as second expression information; an input unit for inputting new third expression information expressing the expression target from any one of the plurality of users; and an integration unit for integrating the second expression information and the third expression information to generate integrated expression information. There are also provided an information processing method using the support apparatus, and a program for causing the support apparatus to operate.

Description

  The present invention relates to a support device, an information processing method, and a program.

Methods for generating subtitles from audio are known (for example, Patent Documents 1 and 2). Also, a method is known in which a subtitle is generated in cooperation with a plurality of workers by causing a worker to create a subtitle and causing another worker to correct the generated subtitle (for example, Patent Document 3). . Also, a method for correcting subtitles by a plurality of workers is known (for example, Patent Document 4). In addition, a method for correcting a subtitle created by automatic speech recognition by an operator is known (for example, Non-Patent Document 1). A method of synthesizing text generated by a plurality of workers is known (for example, Non-Patent Document 2).
[Patent Document 1] JP 2005-228178 [Patent Document 2] JP 2008-32789 [Patent Document 3] JP 2010-157961 [Patent Document 4] JP 2004-226910 [Non-Patent Document 3] [Patent Document 1] Nagatsuma, Fukuda, Yanaginuma, Hirose, “Efficient subtitle creation method using crowdsourcing”, IEICE Technical Report, Vol. 112, no. 336 (2012)
[Non Patent Literature 2] Naim, D.M. Gildea, W.M. Lasecki, J. et al. P. Bigham, “Text Alignment for Real-Time Crown Capturing” The Association for Computational Linguistics, HLT-NAACL, page 201-210 (2013)

  However, if only subtitle correction is repeated continuously, the subtitles are locally optimized and the subtitle quality cannot be sufficiently improved. Further, when subtitle correction is repeatedly performed by a plurality of workers according to the conventional method, it has not been possible to appropriately determine when the subtitle correction is to be finished.

  In the first aspect of the present invention, the support device supports the creation of expression information by a plurality of users, and allows any one of the plurality of users to edit the first expression information that represents the expression target. An editing unit that is acquired as second expression information, an input unit that inputs new third expression information that expresses the expression target from any one of a plurality of users, second expression information, and third expression A support device including a first integration unit that integrates information to generate integrated expression information, an information processing method using the support device, and a program for operating the support device are provided.

  It should be noted that the above summary of the invention does not enumerate all the necessary features of the present invention. In addition, a sub-combination of these feature groups can also be an invention.

The structure of the assistance apparatus 10 of this embodiment is shown. The processing flow of the assistance apparatus 10 of this embodiment is shown. An example of the input screen of the 3rd expression information in this embodiment is shown. An example of the edit screen of the 1st expression information in this embodiment is shown. An example of the expression information integration method in the present embodiment will be described. It is a box-and-whisker figure which shows an example of the effect of the assistance apparatus 10 in this embodiment. It is a box-and-whisker figure which shows an example of the effect of the assistance apparatus 10 in this embodiment. It is a box-and-whisker figure which shows an example of the effect of the assistance apparatus 10 in this embodiment. The processing flow of the assistance apparatus 10 in the modification of this embodiment is shown. 2 shows an example of a hardware configuration of a computer 1900.

  Hereinafter, the present invention will be described through embodiments of the invention, but the following embodiments do not limit the invention according to the claims. In addition, not all the combinations of features described in the embodiments are essential for the solving means of the invention.

  FIG. 1 shows a configuration of a support apparatus 10 according to the present embodiment. The support device 10 supports the work of creating expression information from expression objects by a plurality of users. For example, the support device 10 supports the work of creating expression information that is text such as subtitles representing the contents of audio from content to be expressed including audio such as audio or video. As an example, the support apparatus 10 may support a work of generating a subtitle text that directly represents the content spoken by the person or a text obtained by translating the content spoken by the person as the text representing the content of the voice. The support apparatus 10 includes a dividing unit 102, an automatic recognition unit 104, an editing unit 106, a determination unit 108, an input unit 110, a first integration unit 112, a control unit 114, and a second integration unit 120.

  The dividing unit 102 divides the content to be expressed and generates a plurality of expression targets to be worked. For example, the dividing unit 102 inputs content to be expressed including audio such as video, and divides the content into groups such as sentence units, paragraph units, or context units, and generates a plurality of expression targets. The dividing unit 102 may divide the content to be expressed by detecting breaks in audio and / or changes in the speaker. The dividing unit 102 supplies the generated expression target to the automatic recognition unit 104.

  The automatic recognition unit 104 automatically generates first expression information expressing the expression object from the expression object received from the dividing unit 102. For example, the automatic recognition unit 104 generates text representing the content of speech from speech to be expressed by automatic speech recognition (ASR), which is a conventional technique. The automatic recognition unit 104 supplies the generated text to the editing unit 106 as first expression information.

  The editing unit 106 causes any one of the plurality of users to edit the first expression information, and acquires the edited result as second expression information. Here, one user who edits the first expression information is defined as a first user. The editing unit 106 supplies the first expression information before editing and the second expression information after editing to the determination unit 108, and supplies the second expression information to the first integration unit 112 and the second integration unit 120.

  The determination unit 108 determines whether or not the first expression information is different from the second expression information obtained by editing the first expression information in editing by the editing unit 106. In response to determining that the first expression information and the second expression information are not different, the determination unit 108 causes the editing unit 106 to again present the first expression information to the first user for editing.

  In this re-editing, the determination unit 108 determines whether or not the first expression information and the second expression information coincide with each other in a predetermined number of consecutive edits. In response to determining that the first expression information and the second expression information coincide with each other after a predetermined reference number of consecutive edits, the determination unit 108 stops editing by the editing unit 106 and performs the second integration. This is notified to the unit 120.

  In response to the determination unit 108 determining that the first expression information and the second expression information are different from each other, the input unit 110 transmits new third expression information representing the expression target to any one of the plurality of users. Input from the user. Here, one user who inputs the third expression information is a second user. The input unit 110 supplies the input third expression information to the first integration unit 112 and the second integration unit 120.

  In response to receiving the third expression information, the first integration unit 112 integrates the second expression information and the third expression information to generate integrated expression information. The first integration unit 112 supplies the integrated expression information to the control unit 114.

  The control unit 114 supplies the integrated expression information to the editing unit 106 as new first expression information, causes the editing unit 106 to present and edit the first first expression information to the first user, and separates the first expression information from the first user. To obtain the second expression information. Thereby, the control unit 114 causes the editing process of the expression information by the editing unit 106 to be repeated.

  The second integration unit 120 determines that one or more input first expression information, one or more second expression information, and one or more second expression information are input in response to the end of the repetition of the editing process according to the determination of the determination unit 108. The third expression information is integrated to generate integrated expression information. For example, the second integration unit 120 outputs the integrated expression information as created expression information in response to the determination unit 108 canceling editing by the editing unit 106.

  As described above, the support device 10 acquires the edited second expression information by the editing unit 106, inputs new third expression information by the input unit 110, and the second expression information and the third information by the first integration unit. Repeat the process of integration with expression information. Then, the support device 10 ends the iterative process when it is determined that there is no correction by editing a predetermined reference number of consecutive times, and the first expression information, the second expression information generated so far, And the third expression information is integrated and output.

  Thereby, the support apparatus 10 can determine the timing when the quality of the expression information is improved and the correction no longer occurs as the reference timing for ending the correction of the subtitles and the like. Further, according to the support device 10, when the expression information is corrected, that is, when it is determined that there is room for correction in the expression information, the user is newly input the expression information, thereby preventing local optimization. it can.

  FIG. 2 shows a processing flow of the support apparatus 10 of the present embodiment. In the present embodiment, the support device 10 supports the creation of expression information by a plurality of users by executing the processes of S100 to S170. The support apparatus 10 may execute a processing flow for each of a plurality of expression information generated by dividing the content by the dividing unit 102.

  First, in S110, the automatic recognizing unit 104 automatically generates a text as first expression information from an expression target by ASR (ASR). The automatic recognition unit 104 supplies the generated first expression information to the editing unit 106.

  Next, the support device 10 executes the iterative process related to S120 to S160 indicated by a dotted line in FIG. 2 once or more.

  In S120, the first user edits the text as the first expression information and acquires it as the second expression information (FIX). For example, the editing unit 106 provides the first user with the voice to be expressed, presents the first expression information text to the first user in an editable state, and directly corrects the first expression information to the first user. Then, the text that becomes the second expression information is acquired. As an example, the editing unit 106 supplies the data of the expression target and the text of the first expression information to the information terminal of the first user via the network, and acquires the text of the second expression information from the information terminal of the first user. It's okay.

  In the first iteration, the editing unit 106 presents the first expression information generated by the automatic recognition unit 104 to the first user and acquires the second expression information. In the second and subsequent iterations, the editing unit 106 presents the integrated expression information generated by the first integration unit 112 to the first user as the first expression information, which may be different from the previously input second expression information. The second expression information of is acquired. The editing unit 106 supplies the first expression information before editing and the second expression information after editing to the determination unit 108, and supplies the second expression information to the first integration unit 112 and the second integration unit 120.

  Next, in S130, the determination unit 108 determines whether or not the first expression information is different from the second expression information obtained by editing the first expression information in the previous editing in S120. That is, the determination unit 108 determines whether or not the first expression information has been corrected by editing in the previous S120. If the determination unit 108 determines that the first expression information and the second expression information are different, the determination unit 108 proceeds to S140, and if the determination unit 108 determines that the first expression information and the second expression information are not different The process proceeds to S160.

  In S140, new third expression information expressing the expression target is input from the second user (TYPE). For example, the input unit 110 provides the second user with voice to be expressed and causes the second user to input text that is third expression information corresponding to the expression target. As an example, the input unit 110 may supply data to be expressed to the information terminal of the second user via the network, and obtain the text of the third expression information from the information terminal of the second user.

  In the repeated processing of a plurality of times, the input unit 110 inputs another third expression information that may be different from the previously input third expression information from the second user. The input unit 110 supplies the input third expression information to the first integration unit 112 and the second integration unit 120.

  In S150, the first integration unit 112 integrates the second expression information and the third expression information to generate integrated expression information (MERGE). For example, first, the first integration unit 112 divides the text of the second expression information and the text of the third expression information into a plurality of word elements.

  The first integration unit 112 includes the same word included in the corresponding positions of the text of the second expression information and the third expression information among the elements, and similar words (for example, misspelled existing words, and / Or variant homonyms). For example, the first integration unit 112 may detect words having a left distance within a predetermined threshold as similar words, and associate the words with each other. The 1st integration part 112 arrange | positions all the words containing the word which is not matched on the basis of the matched word.

  When there are two corresponding elements in the text of the second expression information and the text of the third expression information, the first integration unit 112 selects the one that is estimated to be more correct, and the corresponding element does not exist May generate newly integrated text as integrated expression information so as to include the element.

  The first integration unit 112 generates integrated expression information for each repetition process. For example, in each of the plurality of iterative processes, the first integration unit 112 integrates another second expression information that may be different from the previously acquired information and another third expression information that may be different from the previously acquired information. New integrated expression information.

  The first integration unit 112 supplies the generated integrated expression information to the control unit 114. The control unit 114 supplies the integrated expression information to the editing unit 106 as new first expression information, and returns the process to S120. Thereby, in the process of S120 after the 2nd time, control part 114 makes edit part 106 present new 1st expression information to the 1st user, and makes another 2nd expression information acquired.

  In S160, the determination unit 108 determines whether or not the first expression information and the second expression information match in the editing of a predetermined reference number of consecutive times. For example, the determination unit 108 determines whether or not the process of S120 → S130 → S160 has been repeated continuously for a predetermined reference number (n times: for example, twice).

  If the determination unit 108 determines that the first expression information and the second expression information match in the editing of the predetermined number of consecutive reference times, the determination unit 108 proceeds to S170, and if not, the process proceeds to S120. return.

  In S170, the second integration unit 120 integrates the one or more first expression information, the one or more second expression information, and the one or more third expression information that have been input to the editing unit 106 and the input unit 110. , Generate integrated expression information. For example, the second integration unit 120 may integrate all input first expression information, second expression information, and third expression information.

  First, the second integration unit 120 divides each of the plurality of expression information to be integrated into a plurality of elements such as clauses, words, and / or characters, and is common to the corresponding positions of the plurality of expression information. The same elements and similar elements included may be associated with each other. Here, the second integration unit 120 may detect an element having a left distance within a predetermined threshold as a similar element and associate the elements with each other. Further, the second integration unit 120 may perform alignment of the associated elements using a multiple sequence alignment (MSA). Further, the second integration unit 120 may reduce the amount of calculation of the alignment process using the A * algorithm.

  Next, the 2nd integration part 120 may determine the element which should be adopted as final expression information by Majority Vote etc. Here, the second integration unit 120 may give the second expression information a greater weight than the first expression information and the third expression information. Thereby, the support apparatus 10 can reflect the correction result more heavily. The support apparatus 10 outputs the expression information integrated by the second integration unit 120 as created expression information.

  As described above, the support apparatus 10 causes the editing unit 106 to edit the first expression information generated by the automatic recognition unit 104 by the first user in the first iteration, and if there is a correction in editing, the input unit 110 newly adds a third one. The expression information is input, and if there is no correction, the editing unit 106 is caused to edit again. The support device 10 repeats the editing by the editing unit 106 and the input by the input unit 110 until there is no reference correction in the editing for a reference number of times.

  Here, the support apparatus 10 may set the first user and the second user as the same user and / or different users in a plurality of repetitive processes. When the second expression information and / or the third expression information is input from different first users and / or different second users, the support apparatus 10 can be expected to acquire diverse expression information. Further, when the second expression information and / or the third expression information is input from the same first user and / or the same second user, the support apparatus 10 acquires improved expression information each time the repetition proceeds. I can expect that. Moreover, the user who becomes the first user may be the same as or different from the user who becomes the second user.

  Further, the support device 10 may omit the process of S170. In this case, the support apparatus 10 may output the second expression information edited in the last process of S120 as the created expression information.

  In addition, the support apparatus 10 performs the process in S170 in response to the repetition of S120 → S130 → S140 → S150 → S120 being performed in total or continuously for a predetermined reference number (for example, twice) or more. You can proceed to. As a result, the support apparatus 10 can prevent the generation of expression information from being prolonged due to unnecessary editing repetition processing.

  FIG. 3 shows an example of an input screen for third expression information in S140 of the present embodiment. As shown in the figure, the second user reproduces the voice to be expressed together with a moving image or the like on an information terminal or the like, inputs text serving as third expression information corresponding to the expression target, and causes the input unit 110 to acquire the text.

  FIG. 4 shows an example of the editing screen for the first expression information in S120 of the present embodiment. As shown in the figure, the first user reproduces the voice to be expressed together with a moving image or the like on an information terminal or the like, edits the first expression information displayed in advance, creates second expression information, and obtains it in the editing unit 106 Let

  FIG. 5 shows an example of the expression information integration method in S150 of the present embodiment. The FIX line indicates text corresponding to the second expression information received by the first integration unit 112 from the editing unit 106. The TYPE line indicates text corresponding to the third expression information received from the input unit 110 by the first integration unit 112. The MERGE line indicates text corresponding to the integrated expression information generated by the first integration unit 112 by integrating the second expression information and the third expression information.

  As illustrated, the first integration unit 112 divides the text of the second expression information and the third expression information into a plurality of elements, and integrates these elements after alignment. For example, the first integration unit 112 associates the words “It”, “must”, and “suppliing” that are included in common in the second expression information and the third expression information in the same position (first, second, , And 6th).

  The first integration unit 112 may associate a plurality of words corresponding to a variant homophone and / or a spelling error in addition to / instead of associating an element such as a word with the left distance. For example, the first integration unit 112 includes the words “consider” and “consder” and the word “disc” of the variant homophones having the spelling error that is not the variant homophones included in the second representation information and the third representation information. “Disk” may be associated and arranged at the same position (fifth and seventh). The first integration unit 112 arranges the words “now” and “not” that are not associated with the associated word as a reference at different positions (third and fourth).

  Here, the first integration unit 112 generates integrated expression information from the arranged words. For example, the first integration unit 112 directly adopts words (for example, “not” and “now”) that do not have a plurality of corresponding words as integrated expression information. In addition, when there are a plurality of corresponding words, the first integration unit 112 selects one word and adopts it in the integrated expression information.

  For example, when a plurality of corresponding words are heteromorphic homonyms, the first integration unit 112 selects a word related to the third expression information (TYPE). This is because the word related to the second expression information (FIX) may include a word derived from automatic speech recognition (ASR), and it is highly likely that the anomalous homonyms are not contextually correct. It is. As an example, the first integration unit 112 employs “disk” related to the third expression information (TYPE) from “disc” and “disk” that are arranged as the seventh variant homonym.

  In addition, when the plurality of corresponding words are not heteromorphic homonyms, the first integration unit 112 selects a word related to the second expression information (FIX). In such a case, the word related to the third expression information (TYPE) is likely to be a simple input error. As an example, the first integration unit 112 employs “consider” related to the second expression information (FIX) from “consider” and “consder” that is not the fifth homomorphic homonym.

  As a result, the first integration unit 112 generates “It must not know capacitor supplying disk” as integrated expression information from the second expression information (FIX) and the third expression information (TYPE). In this integrated expression information, “not” and “now” are included in duplicate, but the support apparatus 10 can eliminate the overlap by the subsequent repeated processing and the integration processing by the second integration unit 120. . As described above, the support device 10 employs a word related to the third expression information (TYPE) that is not based on automatic speech recognition (ASR) in the case of a variant homonym, otherwise the second expression information (FIX). The support apparatus 10 can generate integrated expression information with higher accuracy by adopting the word related to.

  FIG. 6 is a box-and-whisker diagram illustrating an example of the effect of the support device 10 in the present embodiment. FIG. 6 shows a word error rate (WER) when subtitles are generated by four different methods.

  The leftmost column (ASR) in the figure shows the word error rate when subtitles are generated from speech using only automatic speech recognition (ASR). The second column from the left (CapCap) indicates the word error rate when captions are generated from speech using the support device 10 of the present embodiment. The third column (Type) from the left indicates a word error rate when a subtitle is generated by causing a user to newly input a subtitle (that is, the third expression information generated by the input unit 110 alone). The fourth column (FIX) from the left is a second column in which the subtitle generated from the ASR is corrected once by the user (that is, the first expression information generated by the automatic recognition unit 104 is edited once by the editing unit 106). The word error rate of (expression information) is shown.

  The word error rate is calculated by WER = (I + D + S) / N. Here, I indicates the number of inserted characters necessary for reaching the correct subtitle, D indicates the number of deleted characters necessary for reaching the correct subtitle, S indicates the number of replacement characters required until the correct subtitle is reached, and N indicates Indicates the correct number of subtitle characters.

  The boxes of different sizes in the figure show the WER distribution of the samples included in the range from the first quartile to the third quartile among the WERs of each method. The center line in the rectangle indicates the median WER of the samples included in the rectangle. The-(beard) and + plots are included in the range of 1.5 times the range from the first quartile to the third quartile, and not in the 1.5 times range. The WER of the samples included in the entire data range is shown.

  As shown in the figure, the result (CapCap) by the support device 10 of the present embodiment is the result that the rectangle matches the WER of about 0%, and the WER is the lowest. As a result, according to the support device 10 of the present embodiment, it is apparent that subtitle generation superior to other methods can be supported.

  FIG. 7 is a box-and-whisker diagram showing an example of the effect of the support device 10 in the present embodiment. FIG. 7 shows subtitles (left column) generated only by the ASR stage of S110 and the ASR stage of S110 → FIX of S120 → TYPE of S130 → TYPE S150 when the support apparatus 10 of the present embodiment is used. The subtitles (right column) generated through MERGE are shown. As illustrated, subtitles generated through FIX, TYPE, and MERGE have lower WER than subtitles generated by ASR.

  FIG. 8 is a box-and-whisker diagram showing another example of the effect of the support device 10 in the present embodiment. FIG. 8 shows subtitles (left column) generated through the ASR stage of S110, the FIX of S120, the TYPE of S130, the MERGE of S150, and the ASR of S110 when the support apparatus 10 of the present embodiment is used. The subtitles (right column) generated through the steps FIX of S120, TYPE of S130, MERGE of S150, FIX of S120, TYPE of S130, and MERGE of S150 are shown.

  As shown in FIG. 8, subtitles generated after two repetitions have lower WER than subtitles generated after only one FIX, TYPE, and MERGE repetition. Therefore, it is apparent that the support apparatus 10 of the present embodiment can support generation of sufficiently accurate subtitles by executing FIX, TYPE, and MERGE at least twice. For example, when the repetitive process of S120 → S130 → S140 → S150 has passed a predetermined number of times (for example, twice), the support apparatus 10 may end the repetitive process and proceed to S170.

  FIG. 9 shows a processing flow of the support device 10 in a modification of the present embodiment. In this modification, the editing unit 106 sequentially edits the input first expression information by the first user, acquires the second expression information after editing the reference number, and edits the predetermined reference number. When the determination unit 108 determines that the first expression information and the second expression information are different from each other, the input unit 110 inputs the third expression information.

  Hereinafter, in this modification, the description of the same part as the embodiment described in FIG. 2 may be omitted. For example, the support device 10 according to the present modification may execute the processes of S210, S220, S240, and S270 in the same manner as the processes of S110, S120, S140, and S170 in FIG.

  In S230 of this modification, the determination unit 108 determines whether or not the first expression information is different from the second expression information obtained by editing the first expression information in the previous editing in S220. If the determination unit 108 determines that the first expression information and the second expression information are different, the determination unit 108 advances the process to S232, and determines that the first expression information and the second expression information are not different. The process proceeds to S260.

  In S232, the determination unit 108 determines whether or not the first expression information and the second expression information are different in the continuous editing of a predetermined reference number (m times) in S120. For example, the determination unit 108 determines whether or not the process of proceeding from S220 → S230 → S232 continuously for a predetermined reference number is repeated for the reference number of times. If the result of the determination is affirmative, the determination unit 108 proceeds with the process to S240, and otherwise returns the process to S220.

  In S260, the determination unit 108 determines whether or not the first expression information and the second expression information match in the continuous editing of a predetermined reference number (n times) in the plurality of edits in S220. . For example, the determination unit 108 determines whether or not the process of advancing from S220 → S230 → S260 for a predetermined reference number of times is repeated for the reference number of times. Note that n and m may be the same or different natural numbers. For example, n = 2 and m = 2 may be set.

  If the determination unit 108 determines that the first expression information and the second expression information coincide with each other in the predetermined number of consecutive edits, the process proceeds to S270. If not, the process proceeds to S220. return.

  As described above, the support device 10 according to the present modified example repeats editing by the editing unit 106 and input by the input unit 110 until correction is not performed in editing, and the correction by editing is repeated until the standard number of times occurs. Editing by the editing unit 106 is repeated until the standard number of occurrences continues. Thereby, according to the support device 10 of the present modification example, the second expression information and the third expression information whose quality is further improved through a plurality of continuous edits are integrated, so the expression information to be finally generated The quality can be further improved.

  The support device 10 according to the present embodiment and the modification described so far uses information including sound as an expression target, and includes first expression information, second expression information, subtitles representing the contents of sound as third expression information, and the like. Assisted in generating the text of, but is not limited to this. For example, the support apparatus 10 may use a still image including a photograph, an illustration, a text, and / or a code, and / or a moving image instead of / in addition to the voice as the expression target, and the expression information is the expression target. Corresponding subtitles, translations, and / or musical notes may be used.

  FIG. 10 shows an example of a hardware configuration of a computer 1900 that functions as the support apparatus 10. A computer 1900 according to this embodiment is connected to a CPU peripheral unit having a CPU 2000, a RAM 2020, a graphic controller 2075, and a display device 2080 that are connected to each other by a host controller 2082, and to the host controller 2082 by an input / output controller 2084. Input / output unit having communication interface 2030, hard disk drive 2040, and CD-ROM drive 2060, and legacy input / output unit having ROM 2010, flexible disk drive 2050, and input / output chip 2070 connected to input / output controller 2084 Is provided.

  The host controller 2082 connects the RAM 2020 to the CPU 2000 and the graphic controller 2075 that access the RAM 2020 at a high transfer rate. The CPU 2000 operates based on programs stored in the ROM 2010 and the RAM 2020 and controls each unit. The graphic controller 2075 acquires image data generated by the CPU 2000 or the like on a frame buffer provided in the RAM 2020 and displays it on the display device 2080. Instead of this, the graphic controller 2075 may include a frame buffer for storing image data generated by the CPU 2000 or the like.

  The input / output controller 2084 connects the host controller 2082 to the communication interface 2030, the hard disk drive 2040, and the CD-ROM drive 2060, which are relatively high-speed input / output devices. The communication interface 2030 communicates with other devices via a network by wire or wireless. The communication interface functions as hardware that performs communication. The hard disk drive 2040 stores programs and data used by the CPU 2000 in the computer 1900. The CD-ROM drive 2060 reads a program or data from the CD-ROM 2095 and provides it to the hard disk drive 2040 via the RAM 2020.

  The input / output controller 2084 is connected to the ROM 2010, the flexible disk drive 2050, and the input / output chip 2070, which are relatively low-speed input / output devices. The ROM 2010 stores a boot program that the computer 1900 executes at startup and / or a program that depends on the hardware of the computer 1900. The flexible disk drive 2050 reads a program or data from the flexible disk 2090 and provides it to the hard disk drive 2040 via the RAM 2020. The input / output chip 2070 connects the flexible disk drive 2050 to the input / output controller 2084 and inputs / outputs various input / output devices via, for example, a parallel port, a serial port, a keyboard port, a mouse port, and the like. Connect to controller 2084.

  A program provided to the hard disk drive 2040 via the RAM 2020 is stored in a recording medium such as the flexible disk 2090, the CD-ROM 2095, or an IC card and provided by the user. The program is read from the recording medium, installed in the hard disk drive 2040 in the computer 1900 via the RAM 2020, and executed by the CPU 2000.

  A program that is installed in the computer 1900 and causes the computer 1900 to function as the support device 10 includes a division module, an automatic recognition module, an editing module, a determination module, an input module, a first integration module, a control module, and a first module. 2 integrated modules. These programs or modules work on the CPU 2000 or the like to make the computer 1900 into a dividing unit 102, an automatic recognition unit 104, an editing unit 106, a determination unit 108, an input unit 110, a first integration unit 112, You may function as the control part 114 and the 2nd integration part 120, respectively.

  The information processing described in these programs is read into the computer 1900, whereby the dividing unit 102, which is a specific means in which the software and the various hardware resources described above cooperate, the automatic recognition unit 104, The editing unit 106, the determination unit 108, the input unit 110, the first integration unit 112, the control unit 114, and the second integration unit 120 function. And the specific assistance apparatus 10 according to the intended purpose is constructed | assembled by implement | achieving the calculation or processing of the information according to the intended purpose of the computer 1900 in this embodiment by these specific means.

  As an example, when communication is performed between the computer 1900 and an external device or the like, the CPU 2000 executes a communication program loaded on the RAM 2020 and executes a communication interface based on the processing content described in the communication program. A communication process is instructed to 2030. Under the control of the CPU 2000, the communication interface 2030 reads transmission data stored in a transmission buffer area or the like provided on a storage device such as the RAM 2020, the hard disk drive 2040, the flexible disk 2090, or the CD-ROM 2095, and sends it to the network. The reception data transmitted or received from the network is written into a reception buffer area or the like provided on the storage device. As described above, the communication interface 2030 may transfer transmission / reception data to / from the storage device by a DMA (direct memory access) method. Instead, the CPU 2000 transfers the storage device or the communication interface 2030 as a transfer source. The transmission / reception data may be transferred by reading the data from the data and writing the data to the communication interface 2030 or the storage device of the transfer destination.

  The CPU 2000 is all or necessary from among files or databases stored in an external storage device such as a hard disk drive 2040, a CD-ROM drive 2060 (CD-ROM 2095), and a flexible disk drive 2050 (flexible disk 2090). This portion is read into the RAM 2020 by DMA transfer or the like, and various processes are performed on the data on the RAM 2020. Then, CPU 2000 writes the processed data back to the external storage device by DMA transfer or the like. In such processing, since the RAM 2020 can be regarded as temporarily holding the contents of the external storage device, in the present embodiment, the RAM 2020 and the external storage device are collectively referred to as a memory, a storage unit, or a storage device.

  Various types of information such as various programs, data, tables, and databases in the present embodiment are stored on such a storage device and are subjected to information processing. Note that the CPU 2000 can also store a part of the RAM 2020 in the cache memory and perform reading and writing on the cache memory. Even in such a form, the cache memory bears a part of the function of the RAM 2020. Therefore, in the present embodiment, the cache memory is also included in the RAM 2020, the memory, and / or the storage device unless otherwise indicated. To do.

  In addition, the CPU 2000 performs various operations, such as various operations, information processing, condition determination, information search / replacement, etc., described in the present embodiment, specified for the data read from the RAM 2020 by the instruction sequence of the program. Is written back to the RAM 2020. For example, when performing the condition determination, the CPU 2000 determines whether or not the various variables shown in the present embodiment satisfy the conditions such as large, small, above, below, equal, etc., compared to other variables or constants. If the condition is satisfied (or not satisfied), the program branches to a different instruction sequence or calls a subroutine.

  Further, the CPU 2000 can search for information stored in a file or database in the storage device. For example, in the case where a plurality of entries in which the attribute value of the second attribute is associated with the attribute value of the first attribute are stored in the storage device, the CPU 2000 displays the plurality of entries stored in the storage device. The entry that matches the condition in which the attribute value of the first attribute is specified is retrieved, and the attribute value of the second attribute that is stored in the entry is read, thereby associating with the first attribute that satisfies the predetermined condition The attribute value of the specified second attribute can be obtained.

  The program or module shown above may be stored in an external recording medium. As the recording medium, in addition to the flexible disk 2090 and the CD-ROM 2095, an optical recording medium such as DVD or CD, a magneto-optical recording medium such as MO, a tape medium, a semiconductor memory such as an IC card, and the like can be used. Further, a storage device such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet may be used as a recording medium, and the program may be provided to the computer 1900 via the network.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

  The order of execution of each process such as operations, procedures, steps, and stages in the apparatus, system, program, and method shown in the claims, the description, and the drawings is particularly “before” or “prior to”. It should be noted that the output can be realized in any order unless the output of the previous process is used in the subsequent process. Regarding the operation flow in the claims, the description, and the drawings, even if it is described using “first”, “next”, etc. for convenience, it means that it is essential to carry out in this order. It is not a thing.

  DESCRIPTION OF SYMBOLS 10 Support apparatus, 102 division | segmentation part, 104 automatic recognition part, 106 edit part, 108 judgment part, 110 input part, 112 1st integration part, 114 control part, 120 2nd integration part, 1900 computer, 2000 CPU, 2010 ROM, 2020 RAM, 2030 communication interface, 2040 hard disk drive, 2050 flexible disk drive, 2060 CD-ROM drive, 2070 input / output chip, 2075 graphic controller, 2080 display device, 2082 host controller, 2084 input / output controller, 2090 flexible disk 2095 CD-ROM

Claims (13)

A support device that supports creation of expression information by a plurality of users,
An editing unit that causes any one of a plurality of users to edit the first expression information expressing the expression target, and obtains the second expression information;
An input unit for inputting new third expression information expressing the expression object from any one of a plurality of users;
A first integration unit that integrates the second expression information and the third expression information to generate integrated expression information;
A support device comprising: A controller that supplies the integrated expression information as new first expression information to the editing unit and repeats the processing by the editing unit;
The editing unit receives the new first expression information, presents it to any one of a plurality of users and edits it, and acquires another second expression information from the user.
The support device according to claim 1. The input unit inputs another third expression information from any one of a plurality of users,
The first integration unit generates the integrated expression information by integrating the other second expression information and the other third expression information.
The support device according to claim 2. A determination unit for determining whether the first expression information is different from the second expression information obtained by editing the first expression information;
In response to determining that the first expression information and the second expression information are different from each other, the input unit acquires the third expression information from any one of the plurality of users. ,
The support device according to any one of claims 1 to 3. The editing unit sequentially edits the input first expression information by any one of a plurality of users, acquires the second expression information after editing the reference number of times,
When the determination unit determines that the first expression information and the second expression information are different in editing by the editing unit with a predetermined reference number of times, the input unit inputs the third expression information. To
The support device according to claim 4. In response to the determination unit determining that the first expression information is not different from the second expression information, the editing unit again determines the first expression information as one of the plurality of users. Present and edit to the user
The support device according to claim 4 or 5. The determination unit cancels editing by the editing unit when it is determined that the first expression information and the second expression information coincide with each other in a predetermined number of consecutive second reference number edits. Let
The support device according to any one of claims 4 to 6. An automatic recognition unit that automatically generates first expression information from the expression object;
The editing unit causes a user to edit the first expression information generated by the automatic recognition unit in the first iteration.
The support apparatus according to claim 2 or 3. A second integration unit that integrates one or more input first expression information, one or more second expression information, and one or more third expression information;
Outputting the expression information integrated by the second integration unit as created expression information;
The support device according to claim 1. A division unit that divides the content to be expressed and generates a plurality of the expression objects;
The support apparatus of any one of Claim 1 to 9. The expression object includes sound,
The first expression information, the second expression information, and the third expression information are texts representing the contents of the speech.
The support device according to any one of claims 1 to 10. An information processing method executed by a computer to support creation of expression information by a plurality of users,
An editing stage in which the first expression information expressing the expression object is edited by any one of a plurality of users and acquired as second expression information;
An input step of inputting new third expression information expressing the expression object from any one of a plurality of users;
A first integration step of generating integrated expression information by integrating the second expression information and the third expression information;
An information processing method comprising: A program for supporting creation of expression information by a plurality of users,
When executed on a computer, the computer is
An editing unit that causes any one of a plurality of users to edit the first expression information expressing the expression target, and obtains the second expression information;
An input unit for inputting new third expression information expressing the expression object from any one of a plurality of users;
A first integration unit that integrates the second expression information and the third expression information to generate integrated expression information;
Program to make it work.

Images