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

Abstract

To provide a function which enables efficient character input when a plurality of input means is provided.SOLUTION: A character input device converts an input character string to a corresponding converted character string. The character input device comprises a first input unit, a second input unit, and a character string connection unit. The first input unit accepts input of a first character string in a first input system, The second input unit accepts input of a second character string in a second input system different from the first input system. When accepting input of the second character string following the first character string the input of which has been accepted, the character string connection unit connects the first character string and the second character string, the input of which is currently accepted, to generate a candidate for the input character string. In addition, the character string connection unit extracts a character string matching between a trailing side of the first character string and a leading side of the second character string as a duplicative character string, and when extracting the duplicative character string, generates a character string resulting from connecting the first character string except the duplicative character string and the second character string except the duplicative character string with the duplicative character string therebetween, as one of candidates for the input character string.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a technology for assisting character input.

Background Art

[0002] A user searches for a travel route to a destination using an in-vehicle device (hereinafter, described by taking car navigation as an example). The car navigation has a plurality of input means for inputting a character string corresponding to the destination. For example, this input means is keyboard input using a touch panel and input by voice recognition.

[0003] For example, when the vehicle is stopped (when the traveling speed is 0), the user inputs a character string corresponding to the destination by keyboard using the touch panel of the car navigation.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Patent Document 1 discloses a technology for a character input method having at least two input means such as keyboard input and voice input. For example, when creating an email, the user performs character input using both means of keyboard input and voice input.

[0006] When operating the car navigation using the configuration described in Patent Document 1, the user performs keyboard input while the vehicle is stopped and voice input after the vehicle starts. As a result, the destination of the car navigation becomes a character string (hereinafter, combined character string) obtained by combining the character string input by keyboard and the character string input by voice.

[0007] However, if the combined string is simply a combination of a string entered by the user via keyboard and a string entered via voice input, there is a risk of string duplication due to the different input methods. In other words, the user may not obtain the search results they intended.

[0008] In this case, the user can obtain the desired combined string by intentionally deleting the duplicate strings. However, these operations are very cumbersome and may impair user convenience.

[0009] Therefore, the object of the present invention is to provide a function that enables efficient character input when multiple input means are provided. [Means for solving the problem]

[0010] To achieve the above objective, the character input device of this invention is configured as follows.

[0011] This character input device converts an input string into a corresponding converted string. The character input device comprises a first input unit, a second input unit, and a string concatenation unit. The first input unit accepts input of a first string using a first input method. The second input unit accepts input of a second string using a second input method different from the first input method. When the string concatenation unit receives input of a second string following the first string it has received, it concatenates the first and second strings to generate a candidate input string. The string concatenation unit also extracts matching strings at the end of the first string and the beginning of the second string as duplicate strings. If duplicate strings are extracted, it generates a string as one of the candidate input strings by concatenating the first string (excluding the duplicate strings) and the second string (excluding the duplicate strings) with the duplicate strings.

[0012] In this configuration, even if duplicate characters exist in the concatenated string, the procedure for deleting those duplicate characters can be omitted. In other words, the efficiency of character input for the user can be improved.

[0013] The string concatenation unit of this character input device generates a string by concatenating the second string immediately after the first string, which is then presented as one of the candidate input strings. The character input device also includes a score calculation unit that divides the converted string of each candidate input string into vocabulary units and calculates the co-occurrence frequency between vocabulary units, and a priority determination unit that determines the priority of the converted strings according to the co-occurrence frequency calculated by the score calculation unit.

[0014] The string concatenation unit of this character input device predicts a connecting string that links a first string and a second string, and generates a string formed by connecting the first string and the second string with the connecting string as one of the candidate input strings. The character input device also includes a score calculation unit that divides the converted string of each candidate input string into vocabulary units and calculates the co-occurrence frequency between vocabulary units, and a priority determination unit that determines the priority of the converted strings according to the co-occurrence frequency calculated by the score calculation unit.

[0015] The string concatenation unit of this character input device predicts a connecting string to link the first string and the second string, and generates a string formed by connecting the first string and the second string with the connecting string as one of the candidate input strings.

[0016] In this text input device, either the first or second input method is a key input method, and the other is a voice input method. [Effects of the Invention]

[0017] This invention provides a function that enables efficient character input when multiple input means are provided. [Brief explanation of the drawing]

[0018] [Figure 1] Figure 1 is an illustrative diagram showing how to input text using a text input device as an example of its application. [Figure 2] Figure 2 is a block diagram showing the configuration of the character input device in Configuration Example 1. [Figure 3]FIG. 3 is an image diagram for character input by the character input device of Configuration Example 1. [Figure 4] FIG. 4 is a flowchart showing the processing flow of the character input device of the operation example. [Figure 5] FIG. 5 is a flowchart showing the processing flow of the character input device of Modification Example 1. [Figure 6] FIG. 6 is a block diagram showing the configuration of the character input device of Configuration Example 2. [Figure 7] FIG. 7 is a diagram showing the processing flow of the character input device of Configuration Example 2. [Figure 8] FIG. 8 is a diagram showing the processing flow of the character input device of Configuration Example 2. [Figure 9] FIG. 9 is a flowchart showing the processing flow of the character input device of Configuration Example 2. [Figure 10] FIG. 10 is an image diagram for character input by the character input device of Configuration Example 3. [Figure 11] FIG. 11 is a flowchart showing the processing flow of the character input device of Configuration Example 3. [Figure 12] FIG. 12 is an image diagram for character input by the character input device of Modification Example 2.

Embodiments for Carrying Out the Invention

[0019] Hereinafter, embodiments for carrying out the present invention will be described with reference to several drawings.

[0020] <1. Application Example> FIG. 1 is an image diagram for character input by the character input device 10. The character input device 10 is mounted on, for example, an in-vehicle device. The in-vehicle device is a car navigation that searches for a moving route to the destination by inputting the destination. This in-vehicle device includes the character input device 10 for inputting the destination. The character input device 10 includes a plurality of input means.

[0021] As described above, the text input device 10 is installed in an electronic device such as a car navigation system. However, this electronic device is not limited to a car navigation system; any device that enables text input, such as a tablet or personal computer, is acceptable.

[0022] Let's explain using a specific example of a user entering text. The car navigation system 100 is equipped with a touch panel. The user uses the touch panel to launch an application (hereinafter referred to as "app") that is installed on the car navigation system 100.

[0023] The user launches, for example, a destination search application. The user enters a string of characters into the input display unit 50. Although this example uses a destination search application, any application that has the functionality to allow text input is acceptable, and the type of application is not limited.

[0024] Here, the character input means in the character input device 10 will be described. The character input device 10 includes a first input unit 21 and a second input unit 22. In this invention, the first input unit 21 is keyboard input, and the second input unit 22 is voice input. The first input unit 21 accepts input only when the vehicle is stopped. On the other hand, the second input unit 22 accepts input regardless of whether the vehicle is stopped or in motion.

[0025] A more concrete example will be explained using a scenario where a user sets the string "Kyoto Zoo" and the destination. The user first uses the first input unit 21 for keyboard input, and then uses the second input unit 22 for voice input. The details will be explained below using the configuration shown in Figure 1.

[0026] (1) Character input using the first input unit 21 The user operates the car navigation system while the car is stopped (for example, while waiting at a traffic light). The user uses the first input unit 21 to input the first string "kyoto shi dobutsubu" (Kyoto Animals).

[0027] At this point, the user confirms that the traffic light has turned green and starts the car. As a result, the first input unit 21 stops accepting character input.

[0028] (2) Character input using the second input unit 22 The user confirms that the first input unit 21 is not accepting character input and begins input using the second input unit 22. The user enters the second string "zoo" using the second input unit 22. The user finishes entering the search string by pressing a button, such as the "Execute Search" button.

[0029] (3) Generation of combined strings The character input device 10 combines the first string "kyoto shi dobu" and the second string "doubutsuen" to generate the third string "kyoto shi dobu doubutsuen". At this time, the character input device 10 determines whether there are any duplicate strings (hereinafter referred to as duplicate strings) in the third string.

[0030] The character input device 10 determines that a duplicate string "doubu" exists and deletes the duplicate string "doubu". More specifically, the character input device 10 determines that the string "doubu" at the end of the first string "kyoto to doubu" and the string "doubu" at the beginning of the second string "doubutsuen" are duplicate strings.

[0031] The character input device 10 removes the duplicate character "doubu" from the first string "kyotoshi doubu" and generates "kyotoshi" as part of the first string. Also, the character input device 10 removes the duplicate character "doubu" from the second string "doubutsuen" and generates "tsuen" as part of the second string.

[0032] Next, the character input device 10 combines a part of the first string "kyotoshi", a duplicate string "doubu", and a part of the second string "tsuen" to generate a fourth string "kyotoshi doubutsuen" from "kyotoshi", "doubu", and "tsuen". This fourth string corresponds to the "input string" of the present invention.

[0033] The text input device 10 retrieves the conversion candidate "Kyoto City Zoo" for the fourth string "Kyoto City Zoo" from the dictionary database (hereinafter referred to as the dictionary DB). The text input device 10 outputs the string "Kyoto City Zoo" to the car navigation system. The car navigation system displays location information that matches the string "Kyoto City Zoo".

[0034] This configuration allows users to obtain the intended string even when they combine multiple input methods. Furthermore, if there are duplicate strings in the first and second strings, the system can detect the existence of duplicate strings and delete them.

[0035] In other words, it becomes possible to easily obtain the string of characters intended by the user, allowing users to input text efficiently.

[0036] <2. Configuration Example 1> Figure 2 is a block diagram showing the configuration of the character input device in Configuration Example 1. Figure 3 is an image illustrating character input using the character input device in Configuration Example 1. As mentioned above, the character input device 10 is not limited to car navigation systems; any other electronic device capable of character input is acceptable.

[0037] As shown in Figure 2, the character input device 10 comprises an input unit 20, a detection unit 30, a display control unit 35, a control unit 40, an input display unit 50, and a dictionary DB 60.

[0038] As described above, the input unit 20 includes a first input unit 21 and a second input unit 22. For example, as shown in Figure 1, the first input unit 21 and the second input unit 22 are activated by operating the car navigation system 100.

[0039] The character input device 10 is equipped with a touch panel. This touch panel detects user operations. Operation detection includes, for example, the position of the operation, the duration of the operation, and the temporal change in the position of the operation. The results of these operation detections are output to the detection unit 30. The detection unit 30 outputs these results to the display control unit 35 and the control unit 40 according to the results input from the touch panel. The display control unit 35 outputs to the input display unit 50 to display according to the operation results.

[0040] The first input unit 21 accepts key input, such as from a software keyboard. Various keys for character input are displayed on the first input unit 21. The user inputs a first string of characters using a first input method (key input method) with the first input unit 21.

[0041] The second input unit 22 accepts audio input, such as from a microphone. The second input unit 22 outputs the audio information (second string) spoken by the user to the speech recognition unit (not shown in the diagram). The speech recognition unit obtains feature quantities from the waveform of the audio information and digitizes these feature quantities. Furthermore, the speech recognition unit performs algorithmic analysis on this digitized information and outputs it as a string to the display control unit 35. The display control unit 35 displays the second string on the input display unit 50. In other words, the user inputs the second string using the second input method (audio input method) with the second input unit.

[0042] The control unit 40 comprises a string acquisition unit 41 and a string concatenation unit 42. The control unit 40 is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the character input program according to this invention, the string acquisition unit 41 and the string concatenation unit 42 operate. The memory has an area for loading the character input program according to this invention and an area for temporarily storing data generated during the execution of the character input program. The control unit 40 may be an LSI integrating the hardware CPU, memory, etc. The hardware CPU is a computer that executes the character input method according to this invention.

[0043] The string acquisition unit 41 acquires a first string using the first input method with the first input unit 21, and acquires a second string using the second input method with the second input unit 22. The string acquisition unit 41 outputs the first string and the second string to the string concatenation unit 42.

[0044] The string concatenation unit 42 concatenates the first string and the second string. The detailed procedure will be described later. The string concatenation unit 42 outputs the string obtained by concatenating the first string and the second string to the display control unit 35. The display control unit 35 displays the result on the input display unit 50.

[0045] The user performs a search using the string displayed on the input display unit 50. The character input device 10 searches the dictionary DB (dictionary database) and the map DB (map database) (not shown in the diagram). As a result, the character input device 10 displays the search results and location information.

[0046] Figure 3 will be used to explain the configuration in more detail. The user launches the destination search application of the car navigation system 100. In this example, the user will enter the string "Kyoto City Zoo".

[0047] While the vehicle is stopped, the user uses the first input unit 21 to input the first string "kyoto shi dobu" to the input display unit 50 (state (1) in Figure 3). As described above, the first input unit 21 is a software keyboard. The user inputs the first string "kyoto shi dobu".

[0048] The first input unit 21 outputs the first string "Kyoto Animals" to the detection unit 30. The first input unit 21 outputs the first string "Kyoto Animals" to the detection unit 30. The detection unit 30 outputs the first string "Kyoto Animals" to the display control unit 35 and the string acquisition unit 41. The display control unit 35 displays the first string "Kyoto Animals" on the input display unit 50.

[0049] Next, the user starts the car. As a result, the first input unit 21 stops accepting character input. The user uses the second input unit 22 to input the second character following the first character "kyoto shi dobutsuen" (Kyoto Animals). More specifically, the user uses the second input unit 22 to input the second character "doubutsuen" (Animal Zoo).

[0050] The second input unit 22 outputs the second string "zoo" to the detection unit 30. The first input unit 21 outputs the second string "zoo" to the detection unit 30. The detection unit 30 outputs the second string "zoo" to the string acquisition unit 41 (state (2) in Figure 3).

[0051] The string acquisition unit 41 outputs the first string "kyoto to dōbu" and the second string "dōbutsuen" to the string concatenation unit 42.

[0052] The string concatenation unit 42 combines the first string "kyoto shi dobu" and the second string "doubutsuen" to create the third string "kyoto shi dobu doubutsuen" (state (3) in Figure 3). At this time, the string concatenation unit 42 determines whether the first string "kyoto shi dobu" and the second string "doubutsuen" are duplicate strings.

[0053] In this case, the string concatenation unit 42 determines that the string "doubu" at the end of the first string "kyoto to zoobu" and the string "doubu" at the beginning of the second string "doubutsuen" are duplicate strings.

[0054] The string concatenation unit 42 removes the duplicate string "doubu" from the first string "kyotoshi doubu" and generates "kyotoshi" as part of the first string. Also, the string concatenation unit 42 removes the duplicate string "doubu" from the second string "doubutsuen" and generates "tsuen" as part of the second string.

[0055] Next, the string concatenation unit 42 combines the part of the first string "kyotoshi", the duplicate string "doubu", and the part of the second string "tsuen" to generate the fourth string "kyotoshi doubutsuen". In this way, the string concatenation unit 42 can obtain the conversion candidate "Kyoto City Zoo" intended by the user from the dictionary DB 60.

[0056] In other words, even when a user inputs characters using a combination of multiple input methods (first input unit 21, second input unit 22), the intended string can be obtained. Furthermore, if there are duplicate strings in the first and second strings, the system can determine the existence of duplicate strings and delete them. Therefore, it becomes possible to easily obtain the string intended by the user, enabling the user to input characters efficiently.

[0057] <3. Example of operation> Figure 4 is a flowchart showing the processing flow of the character input device 10 in an example of operation. The processing flow of the character input device 10 will be explained using Figures 1, 2, 3, and 4.

[0058] The first input unit 21 receives the first string entered by the user using the first input method. The first input unit 21 outputs the first string to the detection unit 30. The detection unit 30 outputs the first string to the string acquisition unit 41 (S101).

[0059] Next, the second input unit 22 receives the second string entered by the user using the second input method. The second input unit 22 outputs the second string to the detection unit 30. The detection unit 30 outputs the second string to the string acquisition unit 41 (S102).

[0060] The string acquisition unit 41 outputs the first string and the second string to the string concatenation unit 42. The string concatenation unit 42 concatenates the first string and the second string to create a third string (S103).

[0061] The string concatenation unit 42 determines duplicate strings from the trailing string of the first string and the leading string of the second string. The string concatenation unit 42 removes duplicate strings from the first string and the second string (S104).

[0062] The string concatenation unit 42 creates a fourth string from which duplicate strings have been removed (S105).

[0063] In this way, even when a user inputs characters using a combination of multiple input methods (first input unit 21, second input unit 22), the intended string can be obtained. Furthermore, if there are duplicate strings in the first string and the second string, the duplicate string can be deleted by determining that a duplicate string exists.

[0064] In other words, users can easily obtain the intended string without having to be aware of duplicate strings entered in the first input unit 21 and the second input unit 22. That is, users can input characters efficiently.

[0065] <4. Variation 1> Next, the character input device according to Modification Example 1 will be described with reference to the figure. In Configuration Example 1, only the process of deleting duplicate strings is performed, whereas Modification Example 1 differs in that it also performs the process of determining whether or not duplicate strings exist. The other configurations are the same as those of the character input device 10, and the explanation of the same parts will be omitted.

[0066] Figure 5 is a flowchart showing the processing flow of the character input device 10 in Modification Example 1. After executing the process in step S103 in Configuration Example 1, the string concatenation unit 42 determines whether duplicate strings exist (S110).

[0067] If the string concatenation unit 42 determines that duplicate strings exist (S110: Yes), the string concatenation unit 42 in step S104 executes the process of deleting the duplicate strings starting from the third character.

[0068] On the other hand, if the string concatenation unit 42 determines that a duplicate string exists (S110: No), it creates a fourth string without executing the process in step S104. In other words, the fourth string is identical to the third string. To put it another way, the fourth string is a string created by simply concatenating the first and second strings.

[0069] Even with this configuration, the user can obtain the intended string of characters when they input characters using a combination of multiple input methods (first input unit 21, second input unit 22).

[0070] In other words, it becomes possible to easily obtain the string of characters intended by the user, allowing users to input text efficiently.

[0071] <5. Configuration Example 2> Next, the character input device according to Configuration Example 2 will be described with reference to the figure. Configuration Example 2 differs from Configuration Example 1 in that the control unit 40 of the character input device 10 is equipped with a score calculation unit 43 and a priority determination unit 44. The other components are the same as those of the character input device 10, and the explanation of the similar parts will be omitted.

[0072] Figure 6 is a block diagram showing the configuration of the character input device in Configuration Example 2. Figure 7 is a diagram showing the processing flow of the character input device in Configuration Example 2. Figure 8 is a diagram showing the processing flow of the character input device in Configuration Example 2. Figure 9 is a flowchart showing the processing flow of the character input device in Configuration Example 2.

[0073] The control unit 40 includes a string acquisition unit 41, a string concatenation unit 42, a score calculation unit 43, and a priority determination unit 44. The score calculation unit 43 quantifies the co-occurrence relationships of strings separated by vocabulary. More specifically, when vocabulary A and vocabulary B exist, the score calculation unit 43 calculates the score for when vocabulary A and vocabulary B appear simultaneously.

[0074] (Definition of score calculation) Here, the method by which the score calculation unit 43 calculates the score of the third string will be explained. The string concatenation unit 42 determines whether a string is a duplicate based on the score calculated by the score calculation unit 43. In other words, the score calculation unit 43 generates a score that serves as a criterion for determining whether or not to delete a duplicate string, and the string concatenation unit 42 decides whether or not to delete a duplicate string based on this score. The score is obtained from the sum of the co-occurrence frequencies of adjacent vocabulary.

[0075] The co-occurrence frequency is calculated using co-occurrence status information obtained from sources such as a Japanese co-occurrence dictionary. This co-occurrence frequency is stored in a database for each vocabulary (element) and is saved in dictionary DB60.

[0076] The co-occurrence frequency is defined as follows: a smaller number indicates a higher probability of simultaneous appearance, while a larger number indicates a lower probability. In other words, if the co-occurrence frequency of vocabulary A and vocabulary B is set to "10" and the co-occurrence frequency of vocabulary A and vocabulary C is set to "100", the score calculation unit 43 determines that there is a high probability of simultaneous appearance of vocabulary A and vocabulary B.

[0077] (Method for calculating the score and determining duplicate strings) (1) The string concatenation unit 42 obtains from the dictionary DB 60 candidates for the third string that do not have duplicate strings removed (first candidate) and candidates that have strings removed (second candidate). The string concatenation unit 42 outputs each conversion candidate to the score calculation unit 43. (2) The score calculation unit 43 performs morphological analysis on the first candidate and the second candidate and divides them into vocabulary units. (3) The score calculation unit 43 obtains the co-occurrence frequency of adjacent vocabulary in the first candidate and the second candidate. (4) The score calculation unit 43 calculates the sum (score) of the co-occurrence frequencies of the strings as conversion candidates. (5) The score calculation unit 43 outputs a score for each conversion candidate to the string concatenation unit 42. (6) The string concatenation unit 42 selects the conversion candidate with the smallest score. (7) The string concatenation unit 42 removes duplicate strings according to the score and creates a fourth string.

[0078] Figures 7 and 8 illustrate a more specific example of how the string concatenation unit 42 creates the fourth string.

[0079] Figure 7 shows an example where the first string is "Osaka" and the second string is "Sakai Shitenki". In this case, the string concatenation unit 42 creates "Osaka Sakai Shitenki" and "Osaka Shitenki" as the third string. In this case, the conversion candidate for "Osaka Sakai Shitenki" is designated as the first candidate, and the conversion candidate for "Osaka Shitenki" is designated as the second candidate.

[0080] The string concatenation unit 42 outputs the first candidate "Osaka Sakai City Weather" to the score calculation unit 43. The score calculation unit 43 performs morphological analysis on the first candidate "Osaka Sakai City Weather". In this process, the score calculation unit 43 separates the text into vocabulary items such as "Osaka", "Sakai City", and "Weather". The score calculation unit 43 sets the co-occurrence frequency of "Osaka" / "Sakai City" to "10" and the co-occurrence frequency of "Sakai City" / "Weather" to "10". That is, the score calculation unit 43 calculates the score (first score) of the first candidate to be "20".

[0081] Next, the string concatenation unit 42 outputs the second candidate "Osaka Ishi Tenki" to the score calculation unit 43. The score calculation unit 43 performs morphological analysis on the second candidate "Osaka Ishi Tenki". In this process, the score calculation unit 43 separates the text into vocabulary items such as "Osaka", "Ishi", and "Tenki". The score calculation unit 43 sets the co-occurrence frequency of "Osaka" / "Ishi" to "200" and the co-occurrence frequency of "Ishi" / "Tenki" to "200". That is, the score calculation unit 43 calculates the score (second score) of the second candidate to be "400".

[0082] The score calculation unit 43 compares the first score and the second score and determines that the first score is lower. In other words, the score calculation unit 43 determines that the first candidate "Osaka Sakai City Weather" is the conversion candidate intended by the user. The score calculation unit 43 outputs this result to the priority determination unit 44. The priority determination unit 44 decides to increase the priority of the first candidate "Osaka Sakai City Weather" and outputs this result to the string concatenation unit 42.

[0083] The string concatenation unit 42 outputs the first candidate "Osaka Sakai City Weather" to the display control unit 35. The display control unit 35 increases the priority of the first candidate "Osaka Sakai City Weather" and displays it on the input display unit 50.

[0084] Figure 8 shows an example where the first string is "kyoto shi dobu" and the second string is "doubutsuen". In this case, the string concatenation unit 42 creates "kyoto shi dobu doubutsuen" and "kyoto shi doubutsuen" as the third string. In this case, the conversion candidate for "kyoto shi doubu doubutsuen" is designated as the first candidate, and the conversion candidate for "kyoto shi doubutsuen" is designated as the second candidate.

[0085] The string concatenation unit 42 outputs the first candidate, "Kyoto Guidance Department Zoo," to the score calculation unit 43. The score calculation unit 43 performs morphological analysis on the first candidate, "Kyoto Guidance Department Zoo." In this process, the score calculation unit 43 separates the string into vocabulary words such as "Kyoto," "guidance," "department," and "zoo." The score calculation unit 43 sets the co-occurrence frequency of "Kyoto" / "guidance" to "150," the co-occurrence frequency of "guidance" / "department" to "150," and the co-occurrence frequency of "department" / "zoo" to "200." That is, the score calculation unit 43 calculates the score (first score) of the first candidate to be "500."

[0086] Next, the string concatenation unit 42 outputs the second candidate, "Kyoto City Zoo," to the score calculation unit 43. The score calculation unit 43 performs morphological analysis on the second candidate, "Kyoto City Zoo." In this process, the score calculation unit 43 separates the text into vocabulary items such as "Kyoto City" and "Zoo." The score calculation unit 43 sets the co-occurrence frequency of "Kyoto City" and "Zoo" to "10." That is, the score calculation unit 43 calculates the score for the second candidate (second score) to be "10."

[0087] The score calculation unit 43 compares the first score and the second score and determines that the second score is lower. In other words, the score calculation unit 43 determines that the second candidate, "Kyoto City Zoo," is the conversion candidate desired by the user. The score calculation unit 43 outputs this result to the priority determination unit 44. The priority determination unit 44 decides to increase the priority of the second candidate, "Kyoto City Zoo," and outputs this result to the string concatenation unit 42.

[0088] The string concatenation unit 42 outputs the second candidate, "Kyoto City Zoo," to the display control unit 35. The display control unit 35 prioritizes displaying the second candidate, "Kyoto City Zoo," on the input display unit 50.

[0089] Next, Figure 9 will be used to explain the processing flow of the character input device 10 in Configuration Example 2. Steps S101 to S103 are the same as in Configuration Example 1 and will therefore be omitted.

[0090] The string concatenation unit 42 determines whether duplicate strings exist (S201). If the string concatenation unit 42 determines that no duplicate strings exist (S201: No), it concatenates the first string and the second string to create the fourth string (S202). In other words, the third string is treated as the fourth string.

[0091] The string concatenation unit 42 obtains conversion candidates for the fourth string from the dictionary DB60 (S203).

[0092] On the other hand, if the string concatenation unit 42 determines that duplicate strings exist (S201: Yes), it creates a first candidate and a second candidate (S204). The string concatenation unit 42 outputs the first candidate and the second candidate to the score calculation unit 43.

[0093] The score calculation unit 43 performs morphological analysis on the first and second candidates and separates them into vocabulary units. The score calculation unit 43 also calculates the co-occurrence frequency of each vocabulary unit in the first candidate. Similarly, the score calculation unit 43 calculates the co-occurrence frequency of each vocabulary unit in the second candidate (S205).

[0094] The score calculation unit 43 calculates the sum of the co-occurrence frequencies of the first candidate (first score) and the sum of the co-occurrence frequencies of the second candidate (second score). Based on this, the score calculation unit 43 compares the first score and the second score (S206). The score calculation unit 43 outputs the result of comparing the first score and the second score to the priority determination unit 44.

[0095] The priority determination unit 44 determines the display priority between the first candidate and the second candidate based on the comparison result of the first score and the second score (S207). The score calculation unit 43 outputs the fourth string and the conversion candidates for the fourth string to the string concatenation unit 42.

[0096] By having such a configuration, even when a user inputs characters using a combination of multiple input means (first input unit 21, second input unit 22), the intended string can be obtained.

[0097] Furthermore, the above configuration is not limited to mechanically deleting duplicate strings; it can also determine whether or not to delete duplicate strings based on the vocabulary connections. In other words, it becomes possible to easily obtain the string intended by the user, allowing users to input characters more efficiently.

[0098] <6. Configuration Example 3> Next, the character input device according to Configuration Example 3 will be described with reference to the figure. Configuration Example 3 differs from Configuration Example 2 in that the score determination unit compares a threshold with a score and performs a wildcard search according to the score. The other configurations are the same as those of the character input device 10, and the explanation of the similar parts will be omitted.

[0099] Figure 10 is an illustrative diagram showing character input using the character input device 10 in Configuration Example 3. Figure 11 is a flowchart showing the processing flow of the character input device 10 in Configuration Example 3.

[0100] Figure 10 shows an example where the first string is "osaka" and the second string is "taiikukan". In this case, the string concatenation unit 42 creates "osaka taiikukan" as the third string. The string concatenation unit 42 creates the conversion candidate "Osaka Gymnasium" for the third string "osaka taiikukan".

[0101] The string concatenation unit 42 outputs the conversion candidate "Osaka Gymnasium" to the score calculation unit 43. The score calculation unit 43 performs morphological analysis on the conversion candidate "Osaka Gymnasium". In this process, the score calculation unit 43 separates the text into vocabulary items such as "Osaka" and "Gymnasium". The score calculation unit 43 assigns a co-occurrence frequency of "Osaka" and "Gymnasium" to "200" (first score).

[0102] In this case, the score calculation unit 43 compares the first score with a pre-set threshold. This threshold is an arbitrary value. A low threshold is set if the data exactly matches or closely matches the data stored in the dictionary DB 60, etc. On the other hand, a higher threshold is recommended if the user wants to perform a more flexible search.

[0103] For example, if the threshold is "50", the score calculation unit 43 determines that the conversion candidate is not the string intended by the user because the first score for "Osaka" / "Gymnasium" is "200". In other words, it determines that there is a missing string between the first string "osaka" and the second string "taiikukan". That is, the score calculation unit 43 determines that a connecting string is necessary between the first string "osaka" and the second string "taiikukan". The score calculation unit 43 outputs this result to the string concatenation unit 42.

[0104] The string concatenation unit 42 adds the connecting string "*" between the first string "osaka" and the second string "taiikukan". In other words, the string concatenation unit 42 adds the connecting string as a wildcard. The string concatenation unit 42 makes the third string "osaka*(wildcard)taiikukan" and searches for conversion candidates in the dictionary DB60. The string concatenation unit 42 obtains the first predicted candidate "Osaka Prefectural Gymnasium" and the second predicted candidate "Osaka Municipal Gymnasium".

[0105] The string concatenation unit 42 outputs the first prediction candidate "Osaka Prefectural Gymnasium" and the second prediction candidate "Osaka Municipal Gymnasium" to the score calculation unit 43. The score calculation unit 43 performs morphological analysis on the first prediction candidate "Osaka Prefectural Gymnasium" and the second prediction candidate "Osaka Municipal Gymnasium".

[0106] In this case, the score calculation unit 43 divides the first predicted candidate "Osaka Prefectural Gymnasium" into vocabulary parts such as "Osaka," "Prefectural," and "Gymnasium." The score calculation unit 43 sets the co-occurrence frequency of "Osaka" / "Prefectural" to "10" and the co-occurrence frequency of "Prefectural" / "Gymnasium" to "10." In other words, the score calculation unit 43 calculates the score of the first predicted candidate to be "20."

[0107] Similarly, the score calculation unit 43 divides the second prediction candidate "Osaka Municipal Gymnasium" into vocabulary parts such as "Osaka," "Municipal," and "Gymnasium." The score calculation unit 43 sets the co-occurrence frequency of "Osaka" / "Municipal" to "10" and the co-occurrence frequency of "Municipal" / "Gymnasium" to "15." In other words, the score calculation unit 43 calculates the score of the second prediction candidate to be "25."

[0108] The score calculation unit 43 compares the scores of the first and second prediction candidates with a threshold. As described above, the score calculation unit 43 compares the scores of the first and second prediction candidates with a threshold. The score calculation unit 43 determines that the first and second prediction candidates are the strings intended by the user and outputs these prediction candidates to the string concatenation unit 42.

[0109] The string concatenation unit 42 outputs the first prediction candidate and the second prediction candidate to the display control unit 35. The display control unit 35 displays the first prediction candidate and the second prediction candidate on the input display unit 50. In this case, the display control unit 35 may preferentially display the prediction candidate with the smaller threshold among the first and second prediction candidates.

[0110] Next, Figure 11 will be used to explain the processing flow of the character input device 10 in Configuration Example 3. Steps S101 to S103 and steps S201 and S206 are the same as in Configuration Example 2 and will therefore be omitted.

[0111] If the string concatenation unit 42 determines that no duplicate strings exist (S201: No), it concatenates the first string and the second string to create a fourth string (S202). The string concatenation unit 42 retrieves conversion candidates for the fourth string from the dictionary DB60 (S203).

[0112] The string concatenation unit 42 outputs a conversion candidate for the fourth string to the score calculation unit 43. The score calculation unit 43 determines whether the score is above a threshold (S301).

[0113] If the score calculation unit 43 determines that the score is above a threshold (S301: Yes), it determines that the conversion candidate is not the string intended by the user. The score calculation unit 43 outputs this result to the string concatenation unit 42.

[0114] The string concatenation unit 42 adds a connection string (wildcard) "*" between the first string and the second string, and searches for conversion candidates in the dictionary DB60 (S303). After executing step S303, the string concatenation unit 42 executes step S202 again.

[0115] On the other hand, if the score calculation unit 43 determines that the score is below the threshold (S301: No), it determines that the conversion candidate is the string intended by the user. The score calculation unit 43 confirms the fourth string (S302). The score calculation unit 43 outputs the fourth string and the conversion candidate for the fourth string to the string concatenation unit 42.

[0116] By having such a configuration, even when a user inputs characters using a combination of multiple input means (first input unit 21, second input unit 22), the intended string can be obtained.

[0117] Furthermore, with the above configuration, if it is determined that there is a missing string between the first and second strings, a wildcard search can be performed. In other words, it becomes possible to easily obtain the string intended by the user, allowing the user to input characters more efficiently.

[0118] <7. Variation 2> Next, a character input device according to Modification Example 2 will be described. Modification Example 4 differs from Configuration Example 1 in that the input mode in Configuration Example 1 is a kana input mode, while Modification Example 2 is an English letter input mode. The other configurations are the same as those of the character input device 10, and the explanation of the similar parts will be omitted.

[0119] As shown in Figure 12, the character input device 10 according to Configuration Example 2 differs from the character input device 10 according to Configuration Example 1 in that the first input unit 21 is in kana input mode and the second input unit 22 accepts Japanese input, whereas in Modification Example 2, the first input unit 21 is in English input mode and the second input unit 22 accepts English voice input. The other components of the character input device 10 are the same as those of the character input device 10, and the explanation of the similar parts will be omitted. Note that the first input unit 21 in Figure 12 uses a 12-key keyboard (numeric keypad), but a QWERTY keyboard layout may also be used.

[0120] The following explanation assumes that the first input unit 21 and the second input unit 22 are in English input mode. However, the type of language entered as a string is not limited to English input mode; other types of languages ​​may also be used. For example, the type of language could be Chinese, where the user is asked to input the phonetic transcription of the word to be entered as a string (the desired string), or German, where the user is asked to input the spelling of the word to be entered as a string.

[0121] The first input unit 21 receives the first character, "Kyoto City Z". Next, the second input unit 22 receives the second character, "Zoo".

[0122] The string concatenation unit 42 removes the duplicate character "Z" from the third string "Kyoto City Z Zoo". This creates the fourth string "Kyoto City Zoo".

[0123] In other words, even if a user uses an input mode other than Japanese and combines multiple input methods (first input unit 21, second input unit 22) to input characters, the intended string can be obtained. Furthermore, if there are duplicate strings in the first string and the second string, the system can determine that duplicate strings exist and delete those duplicate strings.

[0124] In other words, it becomes possible to easily obtain the string of characters intended by the user, allowing users to input text efficiently.

[0125] In the example described above, the first input unit 21 was described as character input using a software keyboard, and the second input unit 22 as voice input. However, it is also possible for the first input unit 21 to be voice input and the second input unit 22 to be character input using a software keyboard.

[0126] Furthermore, the first input unit 21 and the second input unit 22 are not limited to text input via a software keyboard or voice input. For example, the first input unit 21 and the second input unit 22 may be handwriting input, gesture input, or the like.

[0127] Furthermore, this invention is not limited to the above example, and in the implementation stage, the components can be modified and materialized without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining the multiple components disclosed in the above example. For example, some components may be deleted from all the components shown in the above example. Furthermore, components from different examples may be appropriately combined.

[0128] Furthermore, the correspondence between the configuration of this invention and the configuration described above can be described as follows: <Note> The character input device (10) converts an input string into a corresponding converted string. The character input device (10) comprises a first input unit (21), a second input unit (22), and a string concatenation unit (42). The first input unit (21) accepts input of a first string using a first input method. The second input unit (22) accepts input of a second string using a second input method different from the first input method. When the string concatenation unit (42) receives input of a second string following the first string it has received, it concatenates the first string and the second string it has received to generate a candidate input string. The string concatenation unit also extracts matching strings at the end of the first string and the beginning of the second string as duplicate strings, and if duplicate strings are extracted, it generates a string as one of the candidate input strings by concatenating the first string (excluding the duplicate strings) and the second string (excluding the duplicate strings) with the duplicate strings. [Explanation of Symbols]

[0129] 10...Text input device 20...Input section 21...First input section 22...Second input section 30...Detection unit 35…Display Control Unit 40... Control Unit 41…Character string acquisition part 42...String concatenation part 43...Score calculation unit 44…Priority determination section 50...Input display section 60…Dictionary DB 100... Car navigation

Claims

1. In a character input device that converts an input string to a corresponding converted string, A first input unit that accepts input of a first string using a first input method, A second input unit that accepts input of a second string using a second input method different from the first input method, The system includes a string concatenation unit that, when a second string is received following the first string that was received as input, concatenates the first string and the second string to generate a candidate input string, The string concatenation part is, At the end of the first string and at the beginning of the second string, matching strings are extracted as duplicate strings. If the duplicate string is extracted, a string is generated as one of the candidate input strings by concatenating the first string (from which the duplicate string has been removed) and the second string (from which the duplicate string has been removed) with the duplicate string. Text input device.

2. The string concatenation unit generates a string by concatenating the second string immediately after the first string, and presents it as one of the candidate input strings. For each candidate input string, a score calculation unit divides the converted string of that input string into vocabulary units and calculates the co-occurrence frequency between vocabulary units. The character input device according to claim 1, further comprising: a priority determination unit that determines the priority order of the converted strings according to the co-occurrence frequency calculated by the score calculation unit.

3. The string concatenation unit predicts a connecting string that joins the first string and the second string, and generates a string formed by joining the first string and the second string with the connecting string as one of the candidate input strings. For each candidate input string, a score calculation unit divides the converted string of that input string into vocabulary units and calculates the co-occurrence frequency between vocabulary units. The character input device according to claim 1, further comprising: a priority determination unit that determines the priority order of the converted strings according to the co-occurrence frequency calculated by the score calculation unit.

4. The character input device according to claim 2, wherein the string concatenation unit predicts a connecting string that connects the first string and the second string, and generates a string formed by connecting the first string and the second string with the connecting string as one of the candidates for the input string.

5. A character input device according to any one of claims 1 to 4, wherein one of the first input method or the second input method is a key input method and the other is a voice input method.

6. A computer in a character input device that converts an input string to a corresponding converted string, The first step involves accepting input of a first string using the first input method, A second step involves accepting input of a second string using a second input method different from the first input method, If a second string is received following the first string that was received as input, a third step is performed to generate a candidate input string by concatenating the first string and the second string that were just received as input. The previous third step is, At the end of the first string and at the beginning of the second string, matching strings are extracted as duplicate strings. If the duplicate string is extracted, a string is generated as one of the candidate input strings by concatenating the first string (from which the duplicate string has been removed) and the second string (from which the duplicate string has been removed) with the duplicate string. The first step is the method of entering text.

7. A computer in a character input device that converts an input string to a corresponding converted string, The first step involves accepting input of a first string using the first input method, A second step involves accepting input of a second string using a second input method different from the first input method, If a second string is received following the first string that was received as input, a third step is performed to generate a candidate input string by concatenating the first and second strings that were just received as input. The previous third step is, At the end of the first string and at the beginning of the second string, matching strings are extracted as duplicate strings. If the duplicate string is extracted, a string is generated as one of the candidate input strings by concatenating the first string (from which the duplicate string has been removed) and the second string (from which the duplicate string has been removed) with the duplicate string. A step-by-step text input program.

Images