JP6058925B2 - Document information reproduction system used for listening test and problem paper used for the document information reproduction system - Google Patents

Description

  The present invention comprises a listening sheet including a problem sheet on which two types of codes are recorded, and a code reading / reproducing device that reads the code recorded on the problem sheet and reproduces information associated with the read code as sound. The present invention relates to a test document information reproduction system and a problem sheet used in the document information reproduction system.

  In the university entrance examination center examination (hereinafter referred to as “center examination”), an English listening test is conducted. The English listening test in the center test will prepare 25 questions. For healthy examinees, an IC player is distributed, and for each question, an English dialogue or sentence is uttered through the IC player for a predetermined number of times (hereinafter referred to as an English utterance through the IC player). Dialogue or text is also called “problem speech”). After the problem voice is played, a certain time is provided during which the voice does not flow. The examinee who heard the question voice reads the question text and the options printed on the question booklet for a certain period of time when the voice does not flow, and marks the answer on the mark sheet. When a certain period of time has passed when no sound flows, the problem sound of the next problem is automatically played through the IC player. That is, the number of times the question sound is played for each question is determined in advance, and the examinee cannot re-listen the question sound beyond the predetermined number of times.

  Currently, in the center examination, from the viewpoint of fairness of the examination environment between healthy examinees and examinees with disabilities, a Braille question booklet with a problem sentence recorded in Braille for students with severe visual impairment, amblyopia For the examinees, an enlarged letter question booklet in which question sentences are recorded in enlarged letters is prepared. The English listening test is conducted in two ways: a continuous method and a sound stop method.

  In the continuous method, the same 25 questions as those of healthy subjects are allocated to a severely visually impaired examinee who uses Braille within the test time 1.5 times that of a healthy examinee. It will be played for a fixed number of times. The examinee listens to the question sound, touches the question sentences and options corresponding to the question sound in the question booklet written in Braille, and writes the answer in Braille. Severely amblyopic test takers are assigned the same 25 questions as normal subjects within 1.3 times the test time of normal takers, and the problem voices are sequentially played through the CD player's earphones a predetermined number of times. The examinee listens to the question sound, reads the question text and the option corresponding to the question sound in the question booklet written in enlarged characters, and answers using a character answer sheet or a general mark sheet. Note that both the severely visually impaired examinee who uses braille and the severely visually impaired examinee who uses enlarged characters have the same problem speech as the normal examinee.

  On the other hand, the sound stop method is a method in which the supervisor operates the CD player according to the examinee's instruction to reproduce the problem sound. Although the question sound can be reproduced according to the examinee's instructions, the test time, the number of questions, and the answer method are the same as in the continuous method.

  In addition, the applicant has proposed a document information reproduction system using document structure template sheets and document structure template sheets for the visually impaired and learning (reading) disabled persons in Patent Document 1. The document structure template sheet described in Patent Document 1 includes only the structure of the test question that has the test subject name, the question number, the paragraph number of the question sentence, the answer number, the option number / symbol, etc. as components. An invisible two-dimensional minimal barcode is recorded in the vicinity of the braille or symbol. A visually impaired person or a learning (reading) person with a disability can systematically grasp the structure of a test question in a short time by using braille or symbols printed on a document structure template sheet. Then, by reading the two-dimensional minimal barcode recorded in the vicinity of the braille or symbol with the barcode reader / player, the detailed information corresponding to the component of the test problem indicated by the braille or symbol is reproduced as sound. Can do.

Japanese Patent No. 4799604

  Candidates who take the English listening test using the sound-stopping method in the center exam can play the problem voice according to their instructions, so they can solve the problem at their own pace according to the speed of reading the question booklet. It becomes. However, one supervisor who operates the CD player is required for each examinee, and when the number of examinees who take the test using the sound-stopping method increases in the future, it is difficult to implement from the viewpoint of securing the supervisor.

  On the other hand, when the English listening test is conducted in a continuous manner, both the visually impaired examinee and the low vision examinee can take the test under the same conditions as the normal examinee. However, both visually impaired examinees and low vision examinees read question booklets more slowly than normal subjects. In particular, visually impaired examinees who read question booklets written in Braille have a speed of reading Braille that is 2 It is about 3 to 3 times slower, and it takes a very long time to read the figure. Therefore, even if the visually impaired examinee is given 1.5 times the test time of the healthy person and the low vision candidate is 1.3 times longer than the healthy person, the time to read the question booklet There is a risk that it may not necessarily be fair with healthy candidates. Also, even among visually impaired examinees and low vision examinees, the speed of reading the question booklet of each individual is different, so there is a limit to the optimization of the test time given to visually impaired examinees, low vision examinees, and healthy examinees. .

  In view of this, it is conceivable for the visually impaired examinee and the low vision examinee to use the document information reproduction system using the document structure template paper described in Patent Document 1 in a continuous manner. Specifically, only the document structure of the question booklet such as the question number, answer number, and option number is printed in braille or symbols, and an invisible two-dimensional minimal barcode is printed in the vicinity of the braille or symbols. The document structure template sheet and the two-dimensional minimal barcode recorded on the document structure template sheet are read, and detailed information (for example, English corresponding to the option number) corresponding to the read two-dimensional minimal barcode is reproduced as audio The code reading / playback device is distributed to visually impaired examinees and low vision examinees. Visually impaired examinees and low vision examinees grasp the document structure of the question booklet using Braille and symbols displayed on the document structure template paper, and code reading and playback device for two-dimensional minimal barcode recorded in the vicinity of Braille and symbols By reading the above, detailed information required by the user can be reproduced as sound. Then, by reproducing the detailed information as audio at a constant speed, the difference between the speed at which the visually impaired examinee or the low vision examinee reads the problem booklet and the speed at which the normal examinee reads the problem booklet can be reduced.

  However, when an English listening test is performed in a continuous manner using the document information reproduction system described in Patent Document 1, it is automatically reproduced at regular intervals through the earphone of the CD player regardless of the examinee's intention. The problem sound overlaps with the sound of the detailed information reproduced at the timing of the examinee by the code reading / reproducing apparatus described in Patent Document 1, and it becomes difficult to accurately hear the problem sound that is reproduced only a predetermined number of times. .

  In view of such circumstances, the present invention is designed to allow examinees who have impaired character recognition, including severely visually handicapped, weakly sighted, and people with developmental disabilities in reading ability, to listen in an examination environment equivalent to that of healthy examinees. It is an object of the present invention to provide a document information reproduction system that makes it possible to take a test.

  According to an aspect of the present invention, there is provided a document information reproducing system used for a listening test, a problem sheet on which a plurality of types of codes are recorded, a reading unit that reads the codes, and details corresponding to the codes and the codes. A storage unit stored in association with the information, a playback unit for reproducing the detailed information stored in the storage unit by voice, and detailed information corresponding to the code read by the reading unit from the storage unit. A code reading / playback device that controls the playback unit and plays back detailed information corresponding to the code read by the reading unit by voice, wherein the code includes a first code and a second code. The control unit determines the type of code read by the reading unit, and determines that the code read by the reading unit is the first code, the control unit The reproduction of the detailed information corresponding to the code is restricted, and when the code read by the reading unit is determined to be the second code, the control unit reproduces the detailed information corresponding to the second code. A document information reproduction system is provided.

  A plurality of types of codes are recorded on the problem paper of the document information reproducing system according to the aspect of the present invention, and the plurality of types of codes include a first code and a second code. Then, the control unit determines whether the code read by the code reading unit is the first code or the second code. If the control unit determines that the code is the first code, the detailed information corresponding to the first code is reproduced. Limit. On the other hand, when it is determined that the code read by the code reading unit is the second code, the detailed information corresponding to the second code is reproduced. Exam questions for the center exam listening test are a question booklet with question sentences and answer options that can be read as many times as the examinee wishes, and are not listed in the question booklet, and playback must be restricted. The question paper of the present invention includes an area indicating that the question voice is reproduced and an area indicating that the question sentence and answer options are reproduced, By recording the first code in the former area and the second code in the latter area, the examinee selects not only the question text and answer options but also the problem voice by his / her own intention. Can be reproduced by the code reading / reproducing apparatus. For this reason, it is possible to avoid a situation in which the voice of the question sentence and the answer choices overlaps with the problem voice and it becomes difficult to accurately listen to any voice. Furthermore, the number of times of the problem voice can be limited. The document information reproducing system according to the aspect of the present invention may further include a third code recording sheet on which a third code readable by the reading unit is printed, and the control unit reads the third code by the reading unit. Unless it is possible, reproduction of detailed information corresponding to the first code may be restricted by prohibiting reproduction of detailed information corresponding to the first code.

  According to the present invention, examinees with impaired character recognition, including severely visually impaired, weakly sighted, and persons with developmental disabilities in reading ability, take a listening test in an examination environment equivalent to that of healthy examinees. Provided is a document information reproduction system that enables the above.

It is a figure which shows the whole structure of a document information reproduction | regeneration system. (A) is a figure which shows a part of problem booklet. (B) is an enlarged view of one cell a1 shown in (a). (C) is an enlarged view of one cell a11 shown in (a). (A) is a schematic block diagram of a code reading / reproducing apparatus. (B) is a figure which shows an example of the information memorize | stored in the memory | storage part of a code reading / reproducing apparatus. It is a figure which shows a part of problem booklet described in the normal character in the English listening problem. FIG. 5 is a diagram showing a state in which the components of the document structure of the problem booklet shown in FIG. 4 are associated with the cells of the template paper. It is a flowchart which shows the flow of the process by the control part of a code reading / reproducing apparatus. It is a figure which shows the specific example of the multimodal problem booklet produced from the problem booklet shown in FIG. It is a table | surface which shows the image of the experiment plan of a 3x3 Greco Latin square method. It is a table | surface which shows the correct answer rate according to a test subject group and a test media. It is a box-and-whisker diagram showing the distribution of the time required for answering by test media. It is a table | surface which shows the estimated value of the test time extension rate according to test subject group and test media.

  Hereinafter, a case where the document information reproducing system according to the present invention is used for an English listening test in a university entrance examination center test will be described as an example. In the following explanation, the “document structure” includes, as constituent elements, the name of the test subject, the major question number, an explanation, an English dialogue or sentence, a sub-question number, a question sentence, an answer number, answer options, etc. It shall mean the structure of the exam questions used in the English listening test.

  First, the configuration of the document information reproduction system of the present invention will be described. As shown in FIG. 1, the document information reproduction system 1 reads a problem booklet 2 in which an invisible code is recorded, and an invisible code recorded in the problem booklet 2, and uses detailed information corresponding to the read code as voice. A code reading / reproducing device 3 to be reproduced and a release code recording sheet 4 on which a release code 41 which is a code for controlling the operation of the code reading / reproducing device 3 is recorded.

  As shown in FIG. 2A, each page of the question booklet 2 includes a template paper P having a plurality of two-dimensionally divided areas a on the surface, and braille 21 printed on the surface of the template paper P. , And an invisible code 22 recorded in the area a where the braille 21 is printed. For convenience of explanation, the outline of the invisible code 22 is indicated by a dotted line in FIGS. 2B and 2C, but this dotted line is not actually printed. Here, “invisible” means that even if characters, numbers, symbols, or diagrams are printed on the code 22 in an overlapping manner, the visual recognition of the characters, numbers, symbols, or diagrams is not hindered.

  The template paper P may be any paper that can print the braille 21 and the invisible code 22. In the present embodiment, as shown in FIG. 2A, on the surface of the template paper P, a plurality of regions a (hereinafter referred to as “cells”) are partitioned in the X direction and the Y direction orthogonal to the X direction. Yes. These cells a are partitioned so that their sizes are equal. In this example, the size of one cell is about 1.0 cm in length and about 0.5 cm in width. In the present embodiment, the template paper P is divided into 40 columns in the X direction and 24 rows in the Y direction. For convenience of explanation, in FIG. 2 (a), FIG. 2 (b), and FIG. 2 (c), the square section is indicated by a solid line, but this solid line is not actually printed.

  On the surface of the template paper P, braille 21 representing one of the elements constituting the document structure is printed. Each Braille 21 is printed so as to fit in one cell a defined on the surface of the template paper P.

  An invisible code 22 called a two-dimensional code, a two-dimensional minimal code, a two-dimensional minimal barcode, or the like is recorded on the cell a of the template paper P on which the braille 21 is printed. The invisible cord 22 is an array of small dots with a spacing of about 0.25 mm, and one cord has a size of 2 mm square. For this reason, as shown in FIGS. 2B and 2C, a plurality of invisible codes 22 including the same information can be printed in one area a of the template paper P. Accordingly, by printing the braille 21 on the invisible code 22, the most invisible codes 22 can be read even when some invisible codes 22 are destroyed, so that high reading reliability can be obtained. Can be obtained.

  The invisible code 22 includes any number from 0 to n, and is classified into a key code 22a (first code) and a normal code 22b (second code) according to the included number. . For example, the key code 22a includes any of numbers 0 to i, and the normal code 22b includes any of numbers i + 1 to n.

  Next, the code reading / reproducing apparatus 3 used with the problem booklet 2 will be described. The code reading / reproducing apparatus 3 includes a storage unit 31, a code reading unit 32, a control unit 33, and a reproducing unit 34 as shown in the schematic configuration diagram of FIG.

  As the storage unit 31, for example, a non-volatile memory or the like can be used. The storage unit 31 stores a number included in the key code 22a or the normal code 22b and detailed information corresponding to the braille 21 printed over the key code 22a or the normal code 22b in association with each other. . In addition to the number and detailed information, the key code 22a is provided with a reproduction prohibition flag (an example of information for limiting the number of reproductions) indicating whether reproduction of the detailed information is prohibited. . When reproducing the detailed information by voice, for example, as shown in FIG. 3B, the file name of the voice file in which the detailed information is recorded is stored in association with the number included in the key code 22a or the normal code 22b. In addition, the audio file itself may be stored in the storage unit 31.

  As the code reading unit 32, any one can be used as long as it can read the key code 22 a, the normal code 22 b recorded in the problem booklet 2, and the release code 41 recorded on the release code recording sheet 4. it can.

  When the code reading unit 32 reads the key code 22a or the normal code 22b recorded in the problem booklet 2, the control unit 33 stores detailed information corresponding to the number based on the number included in the read code. Search from part 31. If the read code is the key code 22a, the reproduction prohibition flag corresponding to the detailed information retrieved from the storage unit 31 is referred to. When the reproduction prohibition flag is set and reproduction of the detailed information is not prohibited, the control unit 33 controls the reproduction unit 34 to reproduce the detailed information, and the reproduction prohibition flag is set and the reproduction of the detailed information is not performed. When it is prohibited, the control unit 33 controls the reproduction unit 34 so as not to reproduce the detailed information. On the other hand, when the read code is the normal code 22b, the control unit 33 controls the reproduction unit 34 to reproduce the detailed information retrieved from the storage unit 31.

  Next, the release code recording sheet 4 (third code recording sheet) on which the release code 41 (third code) is recorded will be described. The release code recording sheet 4 is a sheet provided separately from the problem booklet 2, and the release code 41, which is an invisible code, is recorded in a frame described as “release code” in FIG. . The cancel code 41 is a code for canceling the reproduction prohibition state for all the detailed information stored in the storage unit 31 in association with the key code 22a. When the code reading unit 32 of the code reading / reproducing device 3 reads the release code 41 recorded on the release code recording paper 4, the control unit 33 of the code reading / reproducing device 3 stores the release code 41 in the storage unit 31 in association with the key code 22a. The reproduction prohibition state is canceled by lowering the reproduction prohibition flag for all the stored detailed information. For example, the release code 41 includes a specific number other than 0 to n so that it can be distinguished from the key code 22a and the normal code 22b.

  Next, a method for creating a problem booklet 2 used in the document information reproduction system 1 of the present embodiment from a problem booklet created with normal characters for healthy persons will be described with reference to the drawings. In the following description, “normal character” means a character other than Braille, a number, a symbol, or a figure.

  FIG. 4 is a diagram showing a part of a problem booklet created with normal characters. A part of the question booklet shown in FIG. 4 includes a test subject name, a major question number, an explanatory text about the major question, a minor question number, a question sentence, an answer number, and answer options.

  The creator of the question booklet 2 first becomes a component of the document structure in order to represent the document structure of the first page of the question booklet created with normal characters on the template paper P of the question booklet 2 with Braille 21. The test subject name, major question number, explanatory text about the major question, sub-question number, question sentence, answer number, and answer options are extracted from the question booklet created with normal characters. Then, the extracted components are associated with the cells on the template paper P. Specifically, from the question booklet created with normal characters shown in FIG. 4, the test subject name “English”, the major question number “first question”, the “description” for the major question, and the minor question number “question 1” ], A question sentence, an answer number “1 □”, and answer options “1 to 4” are extracted. In addition, the question speech such as English dialogue sentences or sentences that are circulated by voice a predetermined number of times for each sub-question overlaps with the question sentence and answer choices of question booklet 2 reproduced as speech, both of which are accurate In order to avoid a situation in which it is difficult to hear, the question booklet 2 in the present embodiment also includes an English dialogue sentence or sentence as a component of the document structure. Then, these extracted components are associated with the cells a on the template paper P based on a predetermined rule. For example, in one sub-question, the sub-question number and the English dialogue sentence or sentence are expressed side by side, the question sentence is expressed on the line below it, and the answer number and answer options are further on the line below it. If the document structure of a part of the test question shown in FIG. 4 is expressed on the document structure template sheet 1 based on the rule that the expression is arranged in one line, it is expressed as shown in FIG. In addition, it is necessary to inform the test taker about the method of decoding the document structure created according to the above rules in advance.

  Next, the creator of the question booklet 2 assigns any number from 0 to n to each component of the document structure.

  Then, the creator of the question booklet 2 assigns, for each component of the document structure, an invisible code 22 including any number from 0 to n assigned to the component to the component associated with the component. Print on a. At this time, as described above, since the invisible code 22 is very small, a large number of the same invisible code 22 can be printed on one cell a. Further, the creator of the question booklet 2 prints Braille 21 representing the component of the document structure associated with the cell a on the cell a. That is, the document structure template paper P as shown in FIG. 2A can be created by printing the braille 21 from the invisible code 22 printed on the cell a. The question booklet 2 includes an English conversation sentence or sentence as a component of the document structure, but the English conversation sentence or sentence needs to be reproduced a predetermined number of times. In other words, it is necessary to limit the number of playbacks for English dialogue sentences or sentences. For this reason, the key code 22a including any number from 0 to i is printed on the cells a1 to a6 associated with English conversation sentences or sentences in FIG. FIG. 2 (b) is an enlarged view of one cell a1 in FIG. 2 (a). On the other hand, in FIG. 2A, for example, a normal code 22b including any number i + 1 to n is printed on a cell a11 associated with a component whose number of reproductions is not limited, such as a question sentence. The FIG. 2C is an enlarged view of one cell a11 in FIG. Hereinafter, the problem booklet 2 created in this way is also referred to as a document structure table sound problem booklet.

  Further, the creator of the question booklet 2 allows the examinee to reproduce the necessary information from the question booklet 2 through the code reading / reproducing device 3 as described above. For 31, it is necessary to store the number assigned to the component associated with the cell a on the template paper P and the detailed information of the component in association with each other. Here, the detailed information means specific contents of components such as an explanatory text, an English dialogue text or sentence, a question text, and answer options. Further, when the number of reproductions of the detailed information of the component associated with the cell a is limited, that is, when the key code 22a is printed on the cell a, the reproduction prohibition flag related to the detailed information is displayed. Must be available. For example, in FIG. 5, the number “0” is assigned to the component “dialogue sentence”, and in FIG. 3B, “question 1 dialogue sentence” which is detailed information of the element “dialogue sentence” is configured. It is associated with the number “0” assigned to the element “dialog”. Then, the reproduction prohibition flag relating to “Question 1 dialogue sentence” is in a usable state. On the other hand, when the number of reproductions of the detailed information of the component associated with the cell a is not limited, that is, when the normal code 22b is printed on the cell a, the reproduction prohibition flag is not used. In FIG. 3B, the detailed information in which the reproduction prohibition flag column displays “-” means that the reproduction prohibition flag is not used, that is, the number of reproductions is not limited. As described above, the storage unit 31 of the code reading / reproducing apparatus 3 stores not only the detailed information corresponding to the normal code 22b but also the detailed information corresponding to the key code 22a. Thus, it is possible to reproduce the detailed information in which the number of reproductions is limited and the detailed information in which the number of reproductions is not limited.

  Next, the flow of processing by the control unit 33 of the code reading / reproducing apparatus 3 will be described with reference to the flowchart of FIG. The flowchart in FIG. 6 is based on the premise that the question sheet 2 includes i key codes 22a, and i key codes 22a are represented as k1 to ki.

  First, when the power of the code reading / reproducing apparatus 3 is turned on, the control unit 33 relates to the detailed information stored in the storage unit 31 in association with the numbers included in each of the i key codes k1 to ki. By setting the reproduction prohibition flag, reproduction of detailed information associated with each of the i key codes k1 to ki is prohibited (step S1), and a code reading waiting state is entered.

  Next, the code reading unit 32 reads the key code kj (j is any number from 1 to i) or the normal code 22b from the problem paper 2 or reads the release code 41 from the release code recording paper 4. (Step S2), the process branches to a process flow corresponding to the type of the read code. The type of code is determined based on the number included in each code.

  When it is determined that the code is the normal code 22b based on the number included in the code read by the code reading unit 32, the control unit 33 searches the storage unit 31 for detailed information corresponding to the number, The reproduction unit 34 is controlled to reproduce the detailed information (step S4). After the detailed information is reproduced, the code reading wait state is again entered.

  When the read code is the release code 41, the control unit 33 sets the reproduction prohibition flag for the detailed information stored in the storage unit 31 in association with the numbers included in each of the i key codes k1 to ki. And the reproduction prohibition state of the detailed information associated with each of the i key codes k1 to ki is canceled (step S5). Then, the code reading wait state is entered again.

  After the reproduction prohibition state of the detailed information associated with each of the key codes k1 to ki is canceled in step S5, the code is assigned to i keys based on the number included in the code read by the code reading unit 32. When it is determined that the key code kj is one of the codes, the control unit 33 searches the detailed information stored in the storage unit 31 in association with the number included in the key code kj, and With reference to the reproduction prohibition flag associated with the number, it is determined whether or not reproduction of the detailed information is prohibited (step S6). When the reproduction prohibition state of the detailed information is released (step S6: released), the control unit 33 controls the reproduction unit 34 to reproduce the detailed information (step S7). Then, after reproducing the detailed information, by setting a reproduction prohibition flag, the detailed information is again set in a reproduction prohibited state (step S8), and a code reading waiting state is entered. On the other hand, when the reproduction prohibition state of the detailed information has not been released (step S6: prohibition), the control unit 33 enters the code reading waiting state again without reproducing the detailed information. That is, when the release code 41 is not read by the code reading unit 32 and the key code kj is in a reproduction prohibited state, the control unit 33 prohibits reproduction of detailed information associated with the key code kj. Thus, the reproduction of the detailed information associated with the key code kj is restricted. The detailed information associated with the key code kj is reproduced once in step S7 and is in a reproduction prohibited state by setting a reproduction prohibition flag in step S8. For this reason, when the same key code kj is read again, the control unit 33 does not reproduce the detailed information associated with the key code kj. That is, the number of reproductions of the detailed information is limited to once even if the code reading / reproducing apparatus 3 is turned off and turned on again, as long as the release code 41 is read again and the reproduction prohibition flag is not lowered.

  When the document information reproduction system 1 is used for the listening test of the center test, the test supervisor distributes the question booklet 2 and the code reading / reproducing apparatus 3 to each examinee, but distributes the code reading / reproducing apparatus 3. Before the release code 41 of the release code recording sheet 4 is read, it is necessary to release the reproduction prohibited state of the key code 22a. As a result, the examinee who has received the reading / reproducing apparatus 3 reads the key code 22a recorded in the question booklet 2 with the code reading unit 32, thereby reproducing the detailed information corresponding to the read key code 22a only once. can do.

  The document structure table audio problem booklet described above extracts the components of the document structure from the problem booklet created with normal characters, and associates the extracted components with the cell a on the template paper P based on a predetermined rule. In addition, any number from 0 to n is assigned to each component, including Braille 21 representing the component associated with the cell a and any number from 0 to n assigned to the component It was created by printing the invisible code 22 over the cell a. However, the problem booklet 2 may be created by printing the detailed information of the constituent elements, not the Braille 21 representing the constituent elements, on the template paper P together with the invisible code 22 in normal characters.

  Specifically, a problem booklet created with normal characters is extracted from the components of the document structure, the extracted components are associated with predetermined positions on the template paper P, and each component is assigned one of 0 to n. And the detailed information of the component associated with the position may be printed with the invisible code 22 including the number assigned to the component, superimposed on the position with normal characters. Hereinafter, the problem booklet 2 created in this way is also referred to as a multimodal problem booklet.

  FIG. 7 shows a specific example of the multimodal question booklet created from the question booklet of normal characters shown in FIG. In FIG. 7, the shaded portion means that the invisible code 22 is printed. When the code reading unit 32 of the code reading / reproducing apparatus 3 reads the invisible code 22 printed in the shaded portion, the control unit 33 of the code reading / reproducing apparatus 3 reads the number included in the read code. Based on the information, the storage unit 31 is searched, and the detailed information associated with the number, that is, the content printed in normal characters in the portion indicated by shading is reproduced by the reproduction unit 34. Even when an English listening test is performed using a multi-modal question booklet, the number of playbacks of English dialogue sentences or sentences needs to be limited. For this reason, in FIG. 7, it is necessary to print the key code 22a indicating that the number of reproductions is limited in the shaded portion where the normal character “dialog” is printed. On the other hand, the normal code 22b may be printed on the portion where the detailed information that does not need to limit the number of reproductions, such as a question sentence and answer options, is printed in normal characters.

  In the document information reproduction system 1 of the present invention, an invisible code 22 is recorded in the problem booklet 2, and the invisible code 22 has a number of reproductions of detailed information corresponding to the code limited to a predetermined number. The key code 22a includes reproduction restriction information indicating that the reproduction restriction information is present, and the normal code 22b does not include the reproduction restriction information. The storage unit 31 of the code reading / reproducing apparatus 3 is provided with a reproduction prohibition flag as information for limiting the number of reproductions of detailed information corresponding to the key code 22a. When the code reading unit 32 reads the key code 22a, the control unit 33 refers to the reproduction prohibition flag corresponding to the key code 22a. If the detailed information corresponding to the key code 22a is in the reproduction prohibited state, the detailed information is not reproduced, and if the reproduction prohibited state is released, the detailed information is reproduced. On the other hand, when the code reading unit 32 reads the normal code 22b, the detailed information is reproduced regardless of the number of reproductions of the detailed information corresponding to the normal code 22b. Here, the exam questions for the center exam are based on a question booklet with question texts and answer options that can be read as many times as the examinee intends, and a question voice that is not included in the question booklet. It is configured. Therefore, the problem booklet 2 of the present invention includes an area indicating that the question voice is reproduced and an area indicating that the question sentence and the answer option are reproduced, and the former area includes a key. A code 22a and a normal code 22b are recorded in the latter area. Further, the storage unit 31 of the code reading / reproducing apparatus 3 of the present invention is not limited to the detailed information on the question text and answer options that can be read any number of times, and is not described in the question booklet and the number of times of reproduction is limited. Detailed information (problem voice) is also stored. For this reason, the examinee can select and reproduce not only the question text and the answer choices but also the problem voice by his / her own intention by using one code reading / reproducing apparatus 3. As a result, the voice of the question sentence and answer choices that are selected and played by the examinee's intention, and the problem voice that is automatically reproduced from the IC player distributed in advance to the examinee regardless of the examinee's intention It is possible to avoid a situation in which it is difficult to accurately hear any sound. Furthermore, the number of times of the problem voice can be limited.

  When the document structure table audio question booklet is used as the question booklet, the examinee reads the Braille 21 recorded in the question booklet 2 to systematically grasp the entire structure of the test question in a short time, Necessary information can be actively acquired. In addition, when a multimodal question booklet is used as a question booklet, the examinee can actively read the question booklet by utilizing the characteristics of characters and speech.

  Next, in order to analyze the effect of the test media factors of Braille question booklet, expanded character question booklet, multi-modal question booklet, document structure table speech question booklet and normal character question booklet on the score and answering time of English listening test The evaluation experiment conducted in (1) will be described.

  The experimental design is a 3 × 3 Greco-Latin square method with repetition. FIG. 8 shows an image of the experimental design.

  The factor of the subject group is that the Braille subject group consists of 3 groups of 6 people per group, a total of 18 severely visually impaired high school students. The amblyopic test subject group consists of 9 groups of amblyopic high school students, 3 groups of 3 people each. The healthy subject group consists of 27 groups of healthy high school students, 3 groups of 9 people each.

  There are three test media factors for each group of subjects. The Braille test subject group includes a Braille question booklet (test media A), a real voice document structure table speech problem booklet (test media B), and a synthesized voice document structure table speech problem booklet (test media C). The amblyopic subjects group is an enlarged character problem booklet (test media A), a synthesized speech multimodal problem booklet (test media B), and a synthesized speech document structure table speech problem booklet (test media C). The healthy subject group is a normal text problem booklet (test media A), a real voice document structure table speech problem booklet (test media B), and a synthesized speech document structure table speech problem booklet (test media C).

  The text size and booklet size of the enlarged character problem booklet and multimodal question booklet for the low vision subject group are optimal for the subject from four types of 10.5 points A4, 14 points A4, 18 points A4, and 28 points A3. I was asked to choose something. The character size and booklet size of the document structure table speech problem booklet are 20 points B4.

  The problem booklet and the invisible code were printed using a software grid layouter and LED printer C830 (Oki Data Corporation). The Braille question booklet and the Braille document structure table booklet were printed using a tactile drawing creation system Bplot and a Braille printer / plotter ESA721 (JTR).

  A voice IC player with a two-dimensional code reader, a speaking pen (Grid Mark Co., Ltd.) was used for audio reproduction. All explanations and questions (dialogues and questions) before the start of the listening test are real. The speech playback speed is constant.

  On the other hand, the voice other than the explanation and the problem before the start of the multimodal problem booklet and the document structure table sound problem booklet is a real voice or a synthesized voice. The playback speech speed can be arbitrarily switched to five levels of 0.75 times, 1 times speed, 1.5 times speed, 2 times speed, and 2.5 times speed of the normal speaking speed without changing the pitch.

  The cause of the problem is three sets of English listening test questions for high school entrance exams.

The experimental procedure is a work restriction method that does not limit the test time. It implemented in order of the following (1)-(8).
(1) For each set of problems, the supervisor hands over the subject with an explanation card, a problem card and a problem booklet.
(2) The supervisor turns on the voice IC player, touches the release code to release the voice reproduction prohibition state of the two key codes in the explanation and the problem, and hands them to the subject.
(3) The subject enters the answer start time on the answer sheet while watching the stopwatch. The supervisor replaces and fills in Braille subjects.
(4) The subject touches the explanation card with the voice IC player and listens to the explanation before answering once.
(5) Ask the question booklet choices to be read twice. Since it is difficult to listen to the problem voice of the listening test while reading the Braille question booklet and the two kinds of voice question booklet, after reading the explanation voice, ask the question booklet to be read twice before listening to the problem. Test conditions were also made fair by using the audio problem booklet and the normal text problem booklet. As for the document structure table audio problem booklet, the choices are reproduced twice to be heard.
(6) Touch the question card and listen to the dialogue and questions.
(7) Read the question booklet and ask them to write an answer.
(8) Ask the stopwatch to enter the answer end time on the answer sheet. The supervisor replaces and fills in Braille subjects.

  Since the number of questions by subject group and test media is one question each, the correct answer rate was obtained by dividing the number of correct answers by the number of subjects. The result is shown in FIG. The accuracy rate of the document structure table audio problem booklet of the low vision subjects group was slightly low at 56%, but the accuracy rate of other test media was very high from 78% to 100%.

  As a result of the analysis of variance of scores for each subject group, no significant main effects were observed in the three subject groups with four exceptions: subject group, test media, problem, and question order, with one exception. The exception was a significant main effect on the test media factors of the low vision subjects. As a result of Schiffe's multiple comparison, the accuracy rate of the document structure table speech problem booklet was significantly lower than that of the enlarged character problem booklet. The cause of the low vision subject group is unknown because the number of subjects is as small as nine. In this experiment, the significance level of all tests was 5%.

  As a result of the Mann-Whitney test, there was no significant difference between the Braille question booklet of the Braille subject group, the enlarged character question booklet of the low vision subject group, and the correct rate of the normal character question booklet of the healthy subject group. It is estimated that the level of learning at the English listening test is almost the same.

  In addition, no significant difference was found between the test media scores of the healthy subject group and the disabled subject group, and it was found that this listening test system can measure listening ability with five types of test media. It was done.

  Next, in order to examine the effects of the four factors of test media, subject group, questions, and question order on the time required for answering, we performed analysis of variance and multiple comparison of Schiffe for each subject group.

  The main effect of the test media factors was not significant in the braille group. However, other low vision subjects and healthy subjects were significant.

  The main effects of the subject group factors were significant in all three subject groups.

  The main effect of the problem factor was significant in the low vision group and the healthy subject group. Braille subjects were not significant. However, as a result of Schiffe's multiple comparison, the time required for answering was longer in the order of Set 3, Set 2, and Set 1 for the three subject groups.

  The main effect of the factors in the order of questions was significant in the amblyopic subjects group and the healthy subjects group. Braille subjects were not significant. However, as a result of Schiffe's multiple comparison, the time required for answering became shorter and the order effect was recognized as the questions were placed in the first, second, and third order in the three subject groups.

  Next, in order to compare the distribution of the time required for answering between test media, the distribution of the time required for answering by test media is represented by a box plot shown in FIG. As a result of multiple comparisons of Schiffe, there was no significant difference between the test media connected by the “test result” line.

  The time required for answering the Braille subjects was not significantly significant, although the Braille question booklet was slightly shorter than the other two document structure table speech question booklets. However, the enlarged character problem booklet and the multimodal problem booklet of the low vision subjects group were significantly shorter than the document structure table sound problem booklet. Similarly, the normal text question booklet of the healthy subjects group was significantly shorter than the other two document structure table speech question booklets.

  As a result of the Mann-Whitney test, the test media can be roughly divided into three categories when the time required for answering by test media is compared between test subjects. The normal text question booklet of the healthy subject group is significantly shorter than the Braille question booklet of the Braille subject group, the enlarged character question booklet of the low vision subject group, and the multimodal question booklet. However, there is no significant difference between the Braille question booklet and the enlarged character question booklet or the multimodal question booklet. On the other hand, it was found that the time required for answering the normal text problem booklet, the Braille problem booklet, the expanded character problem booklet, and the multi-modal problem booklet was substantially significantly shorter than that of the voice structure book of the voice structure or synthetic speech.

  Next, divide the lower hinge, median, and upper hinges for the time required for answering the test media in the Braille subjects group and the low vision subject group by the lower hinge, median value, and upper hinge, respectively. Thus, an estimated value of the test time extension rate was obtained. The result is shown in FIG.

  The estimated value of the test time extension from the median was 2.2 times for the Braille subject booklet and 2.6 to 2.9 times for the document structure table speech question booklet in the Braille subjects group. Similarly, in the low vision subject group, the enlarged character problem booklet was 1.6 times, the multimodal problem booklet was 1.9 times, and the document structure table speech problem booklet was 2.0 times.

  As a result of the experiments described above, it has been found that this listening test system can measure listening ability with any of five types of test media. If the level of learning in the English listening test of the subject group is the same level, the Braille question booklet, the expanded character question booklet, the multimodal question booklet, and the document structure table speech question booklet of the disabled subject group under conditions that do not limit the test time The accuracy of the English listening test was similar to that of the normal text question booklet of the healthy subjects group, and no significant difference was observed.

  By optimizing the test time extension rate, persons with disabilities can take the English listening test fairly with healthy candidates. Although the time required for answering the normal text question booklet of the healthy subjects group is significantly short, it is possible to conduct a fair and appropriate listening test by estimating the fair test time extension rate for persons with disabilities. If the test time extension rate is more than twice, a fair test is possible by optimizing the amount of questions.

  In the present embodiment described above, a reproduction prohibition flag is provided for the key code 22a in the storage unit 31 of the code reading / reproducing apparatus 3 as an example of information used to limit the number of reproductions of detailed information. Once the detailed information associated with 22 is reproduced once, the number of reproductions is limited to one by setting a reproduction prohibition flag and setting the reproduction prohibition state. However, the information for limiting the number of times the detailed information is reproduced is not limited to the reproduction prohibition flag. For example, the storage unit 31 may store the number of reproductions instead of the reproduction prohibition flag, and may perform control such that detailed information that has reached a predetermined number of reproductions is not reproduced. In this case, the predetermined number of reproductions can be set to 2 or more.

  In the present embodiment described above, the invisible code 22 includes only one of the numbers 0 to n, but other information may be included instead of the number. For example, the invisible code 22 may include position information obtained by coding the position of the cell a on which the invisible code 22 is recorded on the template paper P using a decimal number or a hexadecimal number. Then, the storage unit 31 may store the position information on the template paper P and the detailed information corresponding to the Braille 21 or normal characters printed at the position in association with each other. In this case, of the invisible codes 22, the key code 22a includes information indicating that the number of reproductions of detailed information corresponding to the Braille 21 or normal characters printed at the position is limited in addition to the position information. (Reproduction restriction information) may be included, and such information may not be included in the normal code 22b. As information indicating that the number of reproductions of detailed information is limited, for example, specific characters and numbers may be included in a form that can be distinguished from position information. The storage unit 31 stores position information and detailed information in association with each other, and a reproduction prohibition flag may be further provided for the key code 22a.

  In the document structure table audio problem booklet of this embodiment, a plurality of the same invisible codes 22 associated with the braille 21 are recorded in the cell a on which the braille 21 is printed. The place where 22 is recorded is not limited to the same cell as the cell a on which the braille 21 is printed. For example, based on a predetermined rule that an invisible code 22 is recorded near one of the upper, lower, left and right squares of the square a printed with braille 21, that is, near the square a printed with braille 21. You may create a question booklet. In this case, the examinee only needs to align the code reading unit 32 with a predetermined vicinity of the printing position of the Braille 21 and cause the code reading unit 32 to read the invisible code 22 recorded in the vicinity while touching the Braille 21. be able to.

  As described above, the document information reproduction system according to the present invention has been described as an example of using the English listening test in the university entrance examination center test. However, the test that can use the document information reproduction system according to the present invention is the university entrance examination center test. Is not limited to the English listening test. The question is composed of a question booklet with contents (for example, question texts and answer options) that can be read as many times as the examinee wants, and a question voice that is not included in the question booklet. In addition, the document information reproduction system according to the present invention can be used for a listening test in which the number of times the problem voice is reproduced is limited.

  Also, by properly using the document structure table audio problem booklet of the present invention and the multimodal problem booklet according to the degree of visual impairment or reading disorder of the examinee, the examinee can obtain necessary information from the problem booklet by his / her own will. It can be actively acquired. Therefore, the document information reproducing system of the present invention can be applied to a so-called speed test type listening test that asks how many questions can be answered and answered correctly within a certain test time.

DESCRIPTION OF SYMBOLS 1 Document information reproduction system 2 Problem booklet 3 Code reading / reproducing apparatus 4 Cancel code recording paper P Template paper 21 Braille 22 Invisible code 22a Key code 22b Normal code 31 Memory | storage part 32 Code reading part 33 Control part 34 Playback part 41 Cancel code

Claims (8)

A document information reproduction system used for a listening test,
A problem paper with multiple types of codes
A reading unit that reads the code; a storage unit that stores the code and detailed information corresponding to the code associated with each other; a reproduction unit that reproduces the detailed information stored in the storage unit by voice; and the reading unit A code reading / reproducing unit including: a controller that retrieves detailed information corresponding to the code read by the reading unit from the storage unit and controls the reproducing unit to reproduce the detailed information corresponding to the code read by the reading unit by voice With the device,
The code includes a first code and a second code,
The control unit determines the type of code read by the reading unit,
When it is determined that the code read by the reading unit is the first code, the control unit restricts reproduction of the detailed information corresponding to the first code,
When it is determined that the code read by the reading unit is the second code, the control unit reproduces detailed information corresponding to the second code. A third code recording sheet on which a third code readable by the reading unit is printed;
The control unit restricts the reproduction of the detailed information corresponding to the first code by prohibiting the reproduction of the detailed information corresponding to the first code unless the third code is read by the reading unit. The document information reproducing system according to claim 1.   After the third code is read by the reading unit, the code recorded on the problem paper is read by the reading unit, and the control unit determines that the code read by the reading unit is the first code 3. The document information reproducing system according to claim 2, wherein the control unit reproduces detailed information corresponding to the first code.   The control unit according to claim 3, wherein when the number of reproductions of the detailed information corresponding to the first code exceeds a predetermined number, the reproduction of the detailed information corresponding to the first code is prohibited. Document information reproduction system. The problem paper is recorded on a template paper having a plurality of two-dimensionally partitioned areas on its surface, Braille printed on the template paper area, and an area where the Braille is printed or an area near it. And the code recorded corresponding to the braille,
The document information reproducing system according to any one of claims 1 to 4, wherein the Braille indicates an element constituting a structure of a test question used in the listening test.   6. The document information reproducing system according to claim 5, wherein the code is an invisible code and includes position information on the template sheet of an area where the braille is printed. The problem paper is a template paper having a plurality of two-dimensionally partitioned areas on its surface, normal characters printed on the template paper area, and areas where the normal characters are printed or areas in the vicinity thereof. And the code recorded corresponding to the normal character,
The normal character indicates an element constituting a structure of a test question used in the listening test, or the detailed information associated with the code recorded corresponding to the normal character. Item 5. The document information reproduction system according to any one of Items 1 to 4. 8. The document information reproducing system according to claim 7, wherein the code is an invisible code, and includes position information on the template sheet of an area where the normal character is printed.

Images