Character input device, character input method, and character input program

Abstract

To improve operability of a user to be applied to a character input by simplifying operation for making a user correct a division error of an input character string.SOLUTION: A first conversion character string obtained by converting an input character string and first reliability of the first conversion character string are input to an input unit from an external apparatus. An extraction unit extracts a second conversion character string obtained by converting the input character string by a conversion unit, and a partial character string different from the first conversion character string. A determination unit determines the first conversion character string as a priority conversion character string when the first reliability is larger than a first threshold. An output unit outputs the priority conversion character string and outputs a partial character string extracted as a correction candidate from a non-priority conversion character string by the extraction unit. A correction unit corrects a conversion candidate to a character string replaced to the selected correction candidate in the case of receiving operation to select the correction candidate.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to a technique for inputting a character string. [Background technology]

[0002] Conventionally, there have been character input devices that allow a user to input a desired character string. For example, Patent Document 1 describes a character input device that allows a character string to be input by voice. As is well known, a character input device that allows a character string to be input by voice has a voice recognition unit that recognizes the user's voice input. The voice recognition unit recognizes the input voice (user's pronunciation) to obtain the reading (phonetic notation) of the user's desired character string (confirmed character string). For example, if the language of the character string input by the user is Japanese, the phonetic notation is a kana character string.

[0003] The character input device performs known kana-kanji conversion on the acquired reading as an input character string, for example, and outputs (displays) candidates for the converted character string (conversion candidates). The user performs an operation to select a conversion candidate that is the desired character string from the output conversion candidates. The character input device confirms the conversion candidate selected by the user as the confirmed character string. The character input device of Patent Document 1 also has a configuration that displays, distinguishing from other parts, portions of the input character string that are determined to be possibly inappropriate due to misrecognition of the input voice, etc. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 180329 / 1983 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the character input device divides the input character string into units such as phrases or words, and performs kana-to-kanji conversion on each divided unit. Therefore, if the character input device incorrectly divides the input character string, the user's desired character string (conversion character string) cannot be obtained as a conversion candidate in the kana-to-kanji conversion. An incorrect division of the input character string occurs when the division of the input character string into phrases or words does not match the division of the user's desired character string into phrases or words.

[0006] The character input device of Patent Document 1 simplifies the operation of allowing a user to check and correct errors in an input string (errors in the speech recognition result), but does not simplify the operation of allowing a user to correct errors in dividing the input string (dividing it into units such as phrases or vocabulary).

[0007] The object of this invention is to provide a technology that simplifies the operation of allowing a user to correct errors in dividing an input string (dividing it into phrases, vocabulary, etc.), thereby improving the user's operability when inputting characters. [Means for solving the problem]

[0008] In order to achieve the above object, the character input device of the present invention is configured as follows.

[0009] An external device is connected to the input unit, and a first converted character string obtained by converting an input character string input by the external device and a first reliability of the first converted character string estimated by the external device are input to the input unit.

[0010] The external device may be, for example, a voice input device that inputs an input string by voice, a handwriting input device that inputs an input string by handwriting, or a character input device that inputs an input string by key operation.

[0011] The conversion unit converts the input character string and obtains a second converted character string. The conversion unit performs, for example, a process related to well-known kana-to-kanji conversion. The input character string may be configured to be input from an external device together with the first converted character string and the first reliability, or may be configured to be estimated based on the first converted character string (for example, a configuration in which the phonetic notation (so-called reading) of the first converted character string is estimated as the input character string).

[0012] The conversion unit divides the input character string into units such as phrases or vocabulary, and performs a known kana-kanji conversion process for each divided unit.

[0013] The extraction unit compares the first converted character string with the second converted character string, and extracts subcharacter strings that differ between these two character strings.

[0014] The determination unit determines one of the first converted character string or the second converted character string as the priority converted character string and the other as the non-priority converted character string. For example, the determination unit may be configured to determine the first converted character string as the priority converted character string when the first reliability is greater than a first threshold. The determination unit may be configured to determine the second converted character string as the priority converted character string when the first reliability is less than a second threshold. The second threshold is less than the first threshold. The conversion unit may acquire a second reliability of the second converted character string, and when the first reliability is less than the first threshold and greater than the second threshold (i.e., when the first reliability is between the first threshold and the second threshold), the determination unit may determine the first converted character string as the priority converted character string if the first reliability is greater than the second reliability, and conversely, determine the second converted character string as the priority converted character string if the second reliability is greater than the first reliability.

[0015] The determination unit may be configured to determine the second converted character string as the priority converted character string when the first reliability is smaller than the first threshold value.

[0016] The output unit outputs the priority conversion character string and outputs the partial character string extracted from the non-priority conversion character string by the extraction unit as a correction candidate. When the correction unit receives an operation to select a correction candidate, it corrects the priority conversion character string to a character string corresponding to the selected correction candidate.

[0017] This allows the user to correct errors in the division of the input character string (division of phrases, vocabulary, etc.) with the simple operation of selecting a correction candidate. In other words, the operation of allowing the user to correct errors in the division of the input character string (division of phrases, vocabulary, etc.) is simplified, and the user's operability in character input is improved.

[0018] The extraction unit may also be configured to separate the second converted character string into segments and extract segments that differ from the first converted character string as partial character strings. [Effects of the Invention]

[0019] According to this invention, the operation of allowing the user to correct errors in dividing an input character string (dividing into phrases, words, etc.) can be simplified, and the user's operability in character input can be improved. [Brief explanation of the drawings]

[0020] [Figure 1] FIG. 1 is a diagram showing a system to which the character input device of this example is applied. [Figure 2] 2A and 2B are diagrams showing examples of a preferred conversion character string and correction candidates displayed on the screen of a display device. [Figure 3] FIG. 10 is a diagram showing an example of a display on the screen of the display device when a correction candidate is selected. [Figure 4] FIG. 4(A) is an example of a screen when the correction candidate "Kyouka" is selected, and FIG. 4(B) is an example of a screen when the conversion candidate "Kyouka" is selected. [Figure 5] FIG. 5(A) is an example of a screen when the correction candidate "Ishani" is selected, and FIG. 5(B) is an example of a screen when the conversion candidate "Ishani" is selected. [Figure 6]1 is a block diagram showing the configuration of a main part of a character input device according to this embodiment; [Figure 7] 10 is a flowchart showing the operation of the character input device of this example. [Figure 8] 10 is a diagram showing an example of display of a priority conversion character string and correction candidates on the screen of a display of a character input device of Modification 1. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0021] Hereinafter, an embodiment of the present invention will be described.

[0022] <1. Application Examples> FIG. 1 is a diagram showing a character input system to which the character input device of this example is applied. The character input system of this example has a character input device 1 and an external device 2, and the character input device 1 is connected to the external device 2. The connection between the character input device 1 and the external device 2 may be wired or wireless. The external device 2 may be, for example, a voice input device that inputs an input character string by voice, a handwriting input device that inputs an input character string by handwriting, or a character input device that inputs an input character string by key operation. Here, a case where the external device 2 is a voice input device will be described as an example.

[0023] The external device 2 performs speech recognition processing on the input speech and acquires the phonetic notation of the input speech as an input character string (for example, a kana character string). The external device 2 divides the input character string into units such as phrases or vocabulary, and performs kana-kanji conversion of the input character string for each divided unit. For example, if the external device 2 performs speech recognition processing and acquires an input character string as follows: "I went to the church." Let's take the example of obtaining the following.

[0024] The external device 2 converts the input character string into, for example, "I went to work today" (The / indicates a division by external device 2.) and split it.

[0025] The external device 2 performs kana-kanji conversion processing, for example, "I went to work today." is acquired as a candidate for the converted character string. The converted character string acquired by the external device 2 performing the kana-kanji conversion process corresponds to the first converted character string referred to in this invention. Hereinafter, the converted character string acquired by the external device 2 performing the kana-kanji conversion process will be referred to as the first converted character string.

[0026] Furthermore, the external device 2 estimates the reliability of the first converted character string acquired by performing the kana-to-kanji conversion process. The reliability of the first converted character string is the likelihood that the first converted character string is the character string desired by the user. The external device 2 estimates the reliability of the first converted character string using a predetermined method. The estimation of the reliability of the first converted character string may be performed, for example, by referring to the kana-to-kanji conversion history and calculating the probability that the character string desired by the user is the first converted character string, or may be performed using other methods.

[0027] The external device 2 outputs the first converted character string "I went to the company today" and the reliability of this first converted character string to the character input device 1.

[0028] In the following description, the reliability of the converted character string that the external device 2 outputs to the character input device 1 will be referred to as a first reliability.

[0029] The character input device 1 acquires the reading (phonetic notation) of the first converted character string input from the external device 2 as an input character string input to the external device 2. In this example, the reading of the first converted character string is a kana character string. The character input device 1 acquires the reading of the first converted character string as an input character string. For example, when the input first converted character string is "I went to my company today," the character input device 1 acquires the reading of the first converted character string (input character string) as follows: "I went to the church." Get.

[0030] The character input device 1 divides the acquired input character string (the reading of the first converted character string) into units of phrases, vocabulary, etc. The result of dividing the input character string in the character input device 1 and the result of dividing the input character string in the external device 2 may match or may differ. For example, the character input device 1 divides the input character string into "I went to the hospital today." The character input device 1 performs kana-kanji conversion processing, for example, "Kyouka / went to the doctor" is acquired as a candidate for the converted character string. The converted character string acquired by the character input device 1 performing the kana-kanji conversion process corresponds to the second converted character string referred to in this invention. Hereinafter, the converted character string acquired by the character input device 1 performing the kana-kanji conversion process will be referred to as the second converted character string.

[0031] If the first converted character string and the second converted character string match, the character input device 1 determines the first converted character string as the preferred converted character string. In this case, the character input device 1 does not determine the second converted character string as the non-preferred converted character string.

[0032] Furthermore, if the first converted character string and the second converted character string do not match, the character input device 1 determines one of the first converted character string or the second converted character string as the priority converted character string, and determines the other as the non-priority converted character string. In this example, if the first reliability of the first converted character string is greater than a predetermined first threshold (for example, the first threshold is 0.9), the character input device 1 determines the first converted character string as the priority converted character string. Conversely, if the first reliability of the first converted character string is not greater than the first threshold, the character input device 1 determines the second converted character string as the priority converted character string.

[0033] The character input device 1 extracts different character strings (hereinafter referred to as partial character strings) between the priority conversion character string and the non-priority conversion character string. In this example, "Kyo-sha" of the first conversion character string and "Kyo-ka-isha" of the second conversion character string are extracted as different partial character strings.

[0034] The character input device 1 outputs a priority conversion character string as a conversion candidate, and outputs a partial character string of a non-priority conversion character string as a correction candidate. For example, as shown in FIGS. 2(A) and (B), the character input device 1 displays the priority conversion character string and the correction candidates on the screen of a display 51. A touch panel 52 that detects a user's input operation is arranged on the screen of the display 51. The character input device 1 allocates an operation key display area 60, a priority conversion character string display area 61, and a correction candidate display area 62 on the screen of the display 51. The operation key display area 60 is an area where operation keys are displayed. The priority conversion character string display area 61 is an area where the priority conversion character string is displayed. The correction candidate display area 62 is an area where correction candidates are displayed.

[0035] Fig. 2(A) is a display example when the preferred conversion character string is "I went to the company today" and the non-preferred conversion character string is "I went to the doctor Kyoka." Fig. 2(B) is a display example when the preferred conversion character string is "I went to the doctor Kyoka" and the non-preferred conversion character string is "I went to the company today."

[0036] When a confirmation key, which is one of the operation keys displayed in the operation key display area 60, is operated, the character input device 1 confirms the preferred conversion character string displayed in the preferred conversion character string display area 61 as a confirmed character string.

[0037] When the character input device 1 accepts an operation to select a correction candidate displayed in the correction candidate display area 62, it displays the selected correction candidate and conversion candidates having the same phonetic spelling as the correction candidate on the screen of the display 51. For example, when the screen of the display 51 is the screen shown in Fig. 2(A), when the character input device 1 accepts an input operation by the user to select "Kyouka", it displays "Kyouka", "Kyouka", "Kyouka", etc. having the same phonetic spelling as the correction candidate ("Kyouka") as conversion candidates in the correction candidate display area 62 (see Fig. 3).

[0038] When the character input device 1 receives an operation to select one of the candidates (correction candidates and conversion candidates) displayed in the correction candidate display area 62, it displays a non-priority conversion character string in which the correction candidate has been replaced with the currently selected candidate in the priority conversion character string display area 61 (see FIGS. 4(A) and (B)). FIG. 4(A) is an example of a screen when the selected candidate is "Kyouka", and FIG. 4(B) is an example of a screen when the selected candidate is "Kyouka". As shown in FIGS. 4(A) and (B), when a candidate is selected, if there are other correction candidates, candidates related to the correction candidate are displayed in the correction candidate display area 62. FIGS. 4(A) and (B) show an example in which candidates related to "Isha ni" are displayed in the correction candidate display area 62.

[0039] 2(A), when the character input device 1 receives an input operation from the user to select "Ishani", it displays "Ishani" as a correction candidate, as well as "Ishani", "Ishani", etc., which have the same phonetic spelling as the correction candidate ("Ishani"), as conversion candidates in the correction candidate display area 62 (see FIG. 5(A)). When the character input device 1 receives an input operation from the user to select "Ishani" on the screen shown in FIG. 5(A), it displays the screen shown in FIG. 5(B) on the display 51.

[0040] In this way, the character input device 1 of this example allows the user to easily correct errors in dividing an input character string (dividing into phrases, words, etc.), thereby improving the user's operability in character input.

[0041] Furthermore, in the character input device 1 of this example, the user may select any correction candidate first in the operation for correcting an error in the division of an input character string (operation for selecting a correction candidate) (in the above example, either "Kyouka" or "Ishani" may be selected first). In other words, the user can correct an error in the division of an input character string without worrying about the order in which the correction candidates displayed in the correction candidate display area 62 are selected.

[0042] <2.Configuration example> 6 is a block diagram showing the configuration of the main parts of the character input device 1 of this example. The character input device 1 includes a control unit 11, a display control unit 12, an operation detection unit 13, an input unit 14, and a dictionary database 15 (dictionary DB 15).

[0043] The control unit 11 controls the operation of each part of the main body of the character input device 1. The control unit 11 has a reading acquisition unit 11a, a conversion unit 11b, an extraction unit 11c, a determination unit 11d, and a correction unit 11e. The reading acquisition unit 11a, the conversion unit 11b, the extraction unit 11c, the determination unit 11d, and the correction unit 11e included in the control unit 11 will be described later.

[0044] The display control unit 12 controls the screen display on the display device 51 .

[0045] The operation detection unit 13 detects an input operation by a user. For example, the operation detection unit 13 detects a pressed position on the touch panel 52 attached to the screen of the display device 51 that is pressed by the user.

[0046] The external device 2 is connected to the input unit 14. The input unit 14 receives as input from the external device 2 a first converted character string and a first reliability of the first converted character string.

[0047] The dictionary DB 15 is a database in which, for each vocabulary, a reading (phonetic notation), a conversion character string, a part of speech, a priority, and the like are associated.

[0048] Next, the reading acquisition unit 11a, conversion unit 11b, extraction unit 11c, determination unit 11d, and correction unit 11e included in the control unit 11 will be described.

[0049] The reading acquisition unit 11a acquires the phonetic notation of the first converted character string input from the external device 2 as the reading of this first converted character string.

[0050] The conversion unit 11b acquires a second converted character string corresponding to the reading acquired by the reading acquisition unit 11a. The second converted character string is a character string obtained by converting the reading acquired by the reading acquisition unit 11a from kana to kanji.

[0051] The extraction unit 11c compares the first converted character string with the second converted character string, and extracts a character string that differs between these two character strings as a subcharacter string.

[0052] When the first converted character string and the second converted character string are different character strings, the determination unit 11d determines one of the first converted character string or the second converted character string to be the priority converted character string, and determines the other to be the non-priority converted character string. If the first reliability of the first converted character string input from the external device 2 is greater than a predetermined first threshold (for example, the first threshold is 0.9), the determination unit 11d determines the first converted character string to be the priority converted character string. In other words, if the first reliability of the first converted character string input from the external device 2 is greater than a predetermined first threshold, the determination unit 11d determines the second converted character string to be the non-priority converted character string.

[0053] Furthermore, in this example, the determination unit 11d determines the second converted character string to be the preferred converted character string if the first reliability of the first converted character string input from the external device 2 is not greater than a predetermined first threshold. In other words, the determination unit 11d determines the first converted character string to be the non-preferred converted character string if the first reliability of the first converted character string input from the external device 2 is not greater than a predetermined first threshold.

[0054] The display control unit 12 displays the priority conversion character string as a conversion candidate on the screen of the display device 51, and also displays the partial character string extracted from the non-priority conversion character string by the extraction unit 11c as a correction candidate. The display control unit 12 corresponds to the output unit referred to in this invention.

[0055] When the correction unit 11e receives an operation to select a correction candidate displayed on the screen of the display 51, the correction unit 11e corrects the character string by replacing the conversion candidate with the selected correction candidate.

[0056] The control unit 11 of the character input device 1 in this example is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the character input program according to the present invention, it operates as a reading acquisition unit 11a, a conversion unit 11b, an extraction unit 11c, a determination unit 11d, and a correction unit 11e. The memory also has an area for expanding the character input program according to the present invention and an area for temporarily storing data generated during execution of the character input program. The control unit 11 may be an LSI that integrates the hardware CPU, memory, and the like. The hardware CPU is also a computer that executes the character input method according to the present invention.

[0057] <3. Example of operation> 7 is a flowchart showing the operation of the character input device of this example. The character input device 1 waits for a first converted character string and a first reliability to be input from the external device 2 (s1). For example, if the external device 2 is a voice input device that inputs an input character string by voice, it performs voice recognition processing on the input voice and acquires a phonetic notation of the input voice as an input character string (for example, a kana character string). Also, for example, if the external device 2 is a handwriting input device that inputs an input character string by handwriting, it acquires the reading of the handwritten input character string as the input character string. Also, for example, if the external device 2 is a character input device that inputs an input character string by key operation, it acquires the reading of the character string input by key operation as the input character string.

[0058] The first converted character string is a character string obtained by, for example, publicly known kana-kanji conversion of the input character string by the external device 2. The external device 2 divides the acquired input character string into phrases or vocabulary (phrases in this example) and performs publicly known kana-kanji conversion for each divided unit.

[0059] The first reliability is the likelihood that the first converted character string is the character string desired by the user, estimated by a predetermined method by the external device 2. The first reliability may be estimated by, for example, a method of calculating the probability that the character string desired by the user is the first converted character string by referring to the kana-kanji conversion history, or by other methods.

[0060] When the character input device 1 receives the first converted character string and the first reliability from the external device 2, the character input device 1 acquires the reading of the input first converted character string as an input character string (s2). The reading acquisition unit 11a performs the process related to s2.

[0061] The character input device 1 divides the input character string acquired in s2 into phrases or vocabulary (phrases in this example) (s3), and performs kana-to-kanji conversion processing for each phrase divided in s3 (s4). The conversion unit 11b performs the processing for s3 and s4.

[0062] The result of dividing the input character string in the character input device 1 does not necessarily match the result of dividing the input character string in the external device 2. In other words, the result of dividing the input character string in the character input device 1 may or may not match the result of dividing the input character string in the external device 2. For example, If the input string is "I went to the church," External device 2 splits the input string into "Today I went to work" and The character input device 1 may divide the input character string into "I went to the hospital / today / I went." There are various factors that cause such a situation, such as a difference between the character conversion program installed in the character input device 1 and the character conversion program installed in the external device 2, a difference in learning between the character input device 1 and the external device 2 by the character conversion process performed up to now, and so on.

[0063] Here, the converted character string obtained by the character input device 1 through the kana-kanji conversion process performed at s4 is referred to as a second converted character string.

[0064] The character input device 1 compares the first converted character string with the second converted character string, and extracts the different character strings as substrings (s5). For example, if the first converted character string is "I went to work today" and the second converted character string is "I went to the doctor Kyoka", "I went to work today" and "I went to the doctor Kyoka" are extracted as substrings. As is clear from the above explanation, "I went to work today" is a substring extracted from the first converted character string, and "I went to the doctor Kyoka" is a substring extracted from the second converted character string. The extraction unit 11c performs the process in s5.

[0065] The character input device 1 determines one of the first converted character string or the second converted character string as a priority converted character string, and determines the other as a non-priority converted character string (s6). The determination unit 11d executes the process of s6. In this example, the character input device 1 determines the priority converted character string based on the first reliability of the first converted character string input from the external device 2. Specifically, if the first reliability of the first converted character string is greater than a predetermined first threshold (for example, the first threshold is 0.9), the character input device 1 determines the first converted character string as the priority converted character string. Conversely, if the first reliability of the first converted character string is not greater than the first threshold, the character input device 1 determines the second converted character string as the priority converted character string.

[0066] If the first converted character string and the second converted character string match and no partial character string is extracted in s5, the character input device 1 determines the first converted character string (or the second converted character string) to be the preferred converted character string (it determines that no non-preferred converted character string exists).

[0067] Furthermore, the character input device 1 determines correction candidates for the priority conversion character string (s7). The correction unit 11e performs the process of s7. If the first conversion character string and the second conversion character string match, the character input device 1 does not perform the process of s7. In s7, the character input device 1 determines, as a correction candidate, a phrase to which a substring extracted from the non-priority conversion character string determined in s6 belongs. For example, if the non-priority conversion character string determined in s6 is the second conversion character string, two words, "Kyou" and "Ishani" (to the doctor) are determined as correction candidates. Furthermore, if the non-priority conversion character string determined in s6 is the first conversion character string, two words, "Kyou" (to the doctor) and "Kaishani" (to the company) are determined as correction candidates.

[0068] The character input device 1 displays the priority conversion character string determined in s6 in the priority conversion character string display area 61 allocated on the screen of the display device 51 (s8). Also, in s8, the character input device 1 displays, for each correction candidate determined in s7, the correction candidate in the correction candidate display area 62 allocated on the screen of the display device 51. In s8, for example, the display control unit 12 controls the screen of the display device 51 to the screen shown in Figs. 2(A) and (B).

[0069] The character input device 1 waits for detection of a confirmation operation or a correction operation by the user (s9, s10). When detecting an operation of the confirmation key displayed in the operation key display area 60 by the user, the character input device 1 determines that a confirmation operation has been performed. Also, when detecting an operation of selecting a correction candidate displayed in the correction candidate display area 62 by the user, the character input device 1 determines that a correction operation has been performed.

[0070] When the character input device 1 determines that the confirmation operation has been performed in s9, it outputs the character string displayed in the priority conversion character string display area 61 at that time as the confirmed character string (s11), and returns to s1.

[0071] When determining that a correction operation has been performed in s10, the character input device 1 displays the currently selected correction candidate and conversion candidates having the same phonetic notation as the currently selected correction candidate in the correction candidate display area 62 (see FIG. 3) (s12), and waits for one of the candidates displayed in the correction candidate display area 62 to be selected (s13).

[0072] When one of the candidates displayed in the correction candidate display area 62 in s12 is selected, the character input device 1 corrects the priority conversion character string displayed in the priority conversion character string display area 61 in accordance with the selected candidate (s14), and returns to s9. In s14, the character input device 1 displays the non-priority conversion character string, which is one of the candidates (correction candidates and conversion candidates) displayed in the correction candidate display area 62 and has been replaced with the currently selected candidate, in the priority conversion character string display area 61 (see FIGS. 4(A) and (B)).

[0073] The correction unit 11e performs the processes in steps s12 to s14.

[0074] In this way, the character input device 1 of this example can correct phrase division errors in an input character string and select conversion candidates with a simple operation of selecting candidates displayed in the correction candidate display area 62. Therefore, the character input device 1 of this example can improve the user's operability in character input.

[0075] Although not specifically explained, when a word of the priority conversion character string displayed in the priority conversion character string display area 61 is selected, the character input device 1 of this example displays conversion candidates having the same pronunciation as the selected word in the correction candidate display area 62. When the user selects one of the conversion candidates displayed in the correction candidate display area 62, the character input device 1 of this example replaces the corresponding word (the currently selected word) of the priority conversion character string displayed in the priority conversion character string display area 61 with the currently selected candidate. In this way, the character input device 1 of this example can correct some of the words of the priority conversion character string determined in s6 to words of a desired character string with a simple operation.

[0076] <4. Modifications> Variation 1 The character input device 1 may display the priority conversion character string in the priority conversion character string display area 61 in a way that makes it easy for the user to check phrase breaks in the priority conversion character string. Specifically, as shown in Fig. 8 , a symbol indicating a phrase break (for example, " / ") may be inserted into the priority conversion character string and displayed in the priority conversion character string display area 61.

[0077] This allows the user to easily check whether or not the segmentation of phrases in the preferred conversion character string needs to be corrected.

[0078] Variation 2 In the above example, the character input device 1 determines either the first converted character string or the second converted character string as the prioritized converted character string depending on whether the first reliability is greater than the first threshold value.

[0079] In the character input device 1 of the second modification, the conversion unit 11b estimates the reliability of the second converted character string (hereinafter referred to as the second reliability) together with the second converted character string. The determination unit 11d compares the first reliability with the second reliability, and determines the character string with the higher reliability as the prioritized converted character string.

[0080] Even with this configuration, the same effects as those of the character input device 1 of the above example can be achieved.

[0081] Variation 3 Furthermore, in the character input device 1, the conversion unit 11b estimates the reliability of the second converted character string (hereinafter referred to as second reliability) together with the second converted character string.

[0082] The determination unit 11d determines the first converted character string as the preferred converted character string if the first reliability is greater than a first threshold. Also, the determination unit 11d determines the second converted character string as the preferred converted character string if the first reliability is less than a second threshold. The second threshold is a value less than the first threshold.

[0083] Furthermore, when the first reliability is smaller than the first threshold and larger than the second threshold, the determination unit 11d compares the first reliability with the second reliability, and determines the character string with the higher reliability as the priority conversion character string.

[0084] Even with this configuration, the same effects as those of the character input device 1 of the above example can be achieved.

[0085] In addition, the determination unit 11d may be configured to determine the second converted character string as the preferred converted character string if the first reliability is smaller than the first threshold value and larger than the second threshold value, and to determine the first converted character string as the preferred converted character string if the second reliability is larger than the first reliability by a predetermined value or more, and to determine the first converted character string as the preferred converted character string if the second reliability is not larger than the first reliability by a predetermined value or more.

[0086] Variation 4 Furthermore, the character input device 1 may be configured so that the user can arbitrarily set the method of determining the preferred conversion character string shown in the example above. In this way, the user can increase the frequency with which phrase separators in the preferred conversion character string match phrase separators in the desired character string, depending on the user's usage environment. In other words, the frequency with which the user performs an operation to correct phrase separators can be reduced. This further improves the user's operability in character input.

[0087] Variation 5 Furthermore, in the above example, the first converted character string and the first reliability are input to the character input device 1 from the external device 2, but the character input device 1 may be configured to input an input character string in addition to the first converted character string and the first reliability from the external device 2. In this case, the character input device 1 can dispense with the reading acquisition unit 11a shown in the above example, and need not execute the process related to s2.

[0088] It should be noted that this invention is not limited to the above-described embodiments as they are, and that the components can be modified and embodied in the implementation stage without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, some components may be omitted from all the components shown in the embodiments. Furthermore, components from different embodiments may be appropriately combined. Furthermore, the order of each step in the flowcharts described in all the above examples is merely an example, and may be appropriately changed within the scope of the present invention.

[0089] Furthermore, the correspondence between the configuration according to the present invention and the configuration according to the above-described embodiment can be described as follows: <Additional Notes> [Appendix 1] an input unit (14) to which an external device (2) is connected and to which a first converted character string obtained by converting an input character string input by the external device (2) and a first reliability of the first converted character string estimated by the external device (2) are input; a conversion unit (11b) that converts the input character string and obtains a second converted character string; an extraction unit (11c) that compares the first converted character string with the second converted character string and extracts a subcharacter string that differs between the two character strings; a determining unit (11d) that determines one of the first converted character string and the second converted character string as a priority converted character string and the other as a non-priority converted character string; an output unit (12) that outputs the priority conversion character string and outputs the partial character string extracted from the non-priority conversion character string by the extraction unit as a correction candidate; a correction unit (11e) that, when receiving an operation to select the correction candidate output by the output unit (12), corrects the prioritized conversion character string in accordance with the selected correction candidate, The determination unit (11d) determines the first converted character string as the priority converted character string when the first reliability is greater than a first threshold value.

[0090] [Appendix 2] the determining unit (11d) determines the second converted character string as the preferred converted character string when the first reliability is smaller than a second threshold value; The character input device (1) according to Supplementary Note 1, wherein the second threshold value is smaller than the first threshold value.

[0091] [Appendix 3] The character input device (1) according to Supplementary Note 1, wherein the determination unit (11d) determines the second converted character string as the priority converted character string when the first reliability is smaller than the first threshold value.

[0092] [Appendix 4] The conversion unit (11b) acquires a second reliability of the second converted character string, The character input device (1) according to Appendix 2, wherein the determination unit (11d) determines the first converted character string as the priority converted character string if the first reliability is smaller than the first threshold value and larger than the second threshold value, and if the first reliability is larger than the second reliability, and conversely, determines the second converted character string as the priority converted character string if the second reliability is larger than the first reliability.

[0093] [Appendix 5] 5. The character input device according to claim 1, wherein the extraction unit separates the second converted character string into segments and extracts segments that differ from the first converted character string as the partial character string.

[0094] [Appendix 6] 6. The character input device (1) according to any one of appendices 1 to 5, wherein the input character string is input to the input unit (14) together with the first converted character string and the first reliability.

[0095] [Appendix 7] 7. The character input device (1) according to any one of appendices 1 to 6, wherein the external device (2) includes a voice recognition unit that recognizes input voice and acquires the input character string. [Explanation of symbols]

[0096] 1...Character input device 2…External device 11...Control unit 11a...Reading acquisition section 11b...Conversion section 11c...Extraction part 11d…Decision section 11e…Correction Department 12...Display control unit 13...Operation detection unit 14...Input section 15...Dictionary database (Dictionary DB)

Claims

1. an input unit to which an external device is connected and to which a first converted character string obtained by converting an input character string input by the external device and a first reliability of the first converted character string estimated by the external device are input; a conversion unit that converts the input character string to obtain a second converted character string; an extracting unit that compares the first converted character string with the second converted character string and extracts a subcharacter string that differs between the two character strings; a determination unit that determines one of the first converted character string and the second converted character string as a priority converted character string and the other as a non-priority converted character string; an output unit that outputs the preferentially converted character string and outputs the partial character string extracted by the extraction unit from the non-preferentially converted character string as a correction candidate; a correction unit that, when receiving an operation to select the correction candidate output by the output unit, corrects the prioritized conversion character string in accordance with the selected correction candidate, The determination unit determines the first converted character string to be the prioritized converted character string when the first reliability is greater than a first threshold.

2. The determination unit determines the second converted character string as the prioritized converted character string when the first reliability is smaller than a second threshold value; The character input device according to claim 1 , wherein the second threshold value is smaller than the first threshold value.

3. The character input device according to claim 1 , wherein the determination unit determines the second converted character string to be the prioritized converted character string when the first reliability is smaller than the first threshold value.

4. The conversion unit obtains a second reliability of the second converted character string; 3. The character input device according to claim 2, wherein, when the first reliability is smaller than the first threshold and larger than the second threshold, if the first reliability is larger than the second reliability, the determination unit determines the first converted character string to be the prioritized converted character string, and conversely, if the second reliability is larger than the first reliability, the determination unit determines the second converted character string to be the prioritized converted character string.

5. 5. The character input device according to claim 1, wherein the extraction section divides the second converted character string into phrases, and extracts, as the partial character string, a phrase that differs from the first converted character string.

6. The character input device according to claim 1 , wherein the input unit receives the input string together with the first converted string and the first reliability.

7. 5. The character input device according to claim 1, wherein the external device includes a voice recognition unit that recognizes an input voice and acquires the input character string.

8. a conversion step of converting an input string into a first converted string obtained by converting an input string inputted to the external device and a first reliability of the first converted string estimated by the external device, when the first converted string is inputted from the external device connected to the input unit, and converting the input string to obtain a second converted string; an extraction step of comparing the first converted string with the second converted string and extracting substrings that differ between the two strings; a determining step of determining one of the first converted character string and the second converted character string as a preferred converted character string and the other as a non-preferred converted character string; an output step of outputting the preferentially converted character string and outputting the partial character string extracted from the non-preferentially converted character string in the extraction step as a correction candidate; a correction step of correcting the prioritized converted character string in accordance with the selected correction candidate when an operation of selecting the correction candidate output in the output step is received, The character input method, wherein the determining step is a step of determining the first converted character string as the prioritized converted character string when the first reliability is greater than a first threshold value.

9. a conversion step of converting an input string into a first converted string obtained by converting an input string inputted to the external device and a first reliability of the first converted string estimated by the external device, when the first converted string is inputted from the external device connected to the input unit, and converting the input string to obtain a second converted string; an extraction step of comparing the first converted string with the second converted string and extracting substrings that differ between the two strings; a determining step of determining one of the first converted character string and the second converted character string as a preferred converted character string and the other as a non-preferred converted character string; an output step of outputting the preferentially converted character string and outputting the partial character string extracted from the non-preferentially converted character string in the extraction step as a correction candidate; a correction step of correcting the prioritized converted character string in accordance with the selected correction candidate when an operation of selecting the correction candidate output in the output step is received; The character input program, wherein the determining step is a step of determining the first converted character string as the prioritized converted character string when the first reliability is greater than a first threshold value.

Images