JP2007072671A - Portable information processing device - Google Patents

Abstract

Provided is a portable information processing apparatus that narrows a range of candidates for speech recognition with a simple button operation and improves input efficiency.
SOLUTION: Matched input data corresponding to index data that matches recognized input data recognized by a speech recognition device is placed in a row and a column where the pressed operation button is located among the operation buttons arranged in a matrix form. Based on the input data extracted from the dictionary data of the classification corresponding to the attribute of the recognized input data, the matched input data corresponding to the index data matching the recognized input data is extracted. And when the type of processing based on the determined recognition input data is output, the attribute data corresponding to the attribute of the recognition input data is searched for attribute data equal to the matching input data in the database. Information processing means having attribute data output means for extracting and outputting attribute data of another attribute associated with the attribute data Location.
[Selection] Figure 4

Description

  The present invention relates to a portable information processing apparatus, and in particular, a portable information processing apparatus that simplifies input operations and processing of input information by performing voice recognition by specifying a dictionary classified according to the purpose of information processing by button operation. About.

  In recent years, various information processing functions have been installed in mobile phones in addition to voice communication functions, and have an aspect as an information processing apparatus. Therefore, information input / output to / from such mobile phones is also complicated. For example, a mobile phone equipped with a schedule function needs to appropriately input and output various information such as time, place, and schedule contents.

  When inputting information, it is common to assign a character to an operation button and input a character in response to the button being pressed. However, as the amount of information increases, the operation becomes complicated when trying to input all characters in characters. In addition, an input method based on voice recognition that uses the original voice input / output function to recognize the voice spoken by the user and convert it into a character string has been proposed, but the voice spoken by the user is recognized with high accuracy. For this purpose, it is necessary to perform enormous information processing, which exceeds the processing capability of an information processing device that can be mounted on the casing of a mobile phone.

Therefore, an input method has been proposed in which character input and voice recognition are combined and the accuracy of voice recognition is compensated by button operation. For example, Patent Document 1 and Patent Document 2 describe a method in which a character at the beginning of an input word is input by a button operation, and speech recognition is performed using a word starting from the input character as a candidate.
JP-A-11-17813 JP 2001-255986 A

  However, the conventional technology has the following problems. That is, since character input by button operation is targeted for 50 sounds, if there are a large number of word candidates starting from the input character, the possibility of erroneous recognition is high and the processing efficiency of speech recognition is not improved. Eventually, if the second and subsequent characters are input in order to further narrow the input candidate range, the operation burden on the user is doubled.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a portable information processing apparatus that narrows down the range of candidates for speech recognition with a simple button operation and makes input more efficient.

  In order to achieve the above object, according to the first aspect of the present invention, in response to pressing of an operation button and voice input, a coincidence input that matches the input voice with the recognition input data recognized by the voice recognition device In a portable information processing apparatus that extracts data from dictionary data and performs information processing corresponding to an attribute of the matched input data based on the matched input data, the information processing includes a plurality of information corresponding to the matched input data attribute. And determining the attribute of the coincidence input data based on the position of the pressed operation button among a plurality of operation buttons, and from the dictionary data corresponding to the determined attribute, the recognition input It has a match input data extraction means for extracting match input data that matches the data.

  According to the first aspect, the recognition input data candidates refer to dictionary data classified for each attribute according to information processing, so that it is possible to extract matching input data from a narrower range, Good speech recognition is possible.

  According to a preferred embodiment in the first aspect, the plurality of operation buttons are arranged in a matrix, and the coincidence input data extracting means is configured to match the coincidence based on a row or a column in which the pressed button is located. It is characterized by further determining the type of information processing based on the input data. Therefore, the dictionary selection according to the purpose of information processing can be performed intuitively by associating the classification of the dictionary data and the type of processing based on the recognition input data with the rows and columns of the operation buttons arranged two-dimensionally. In addition, selection of different two-dimensional options such as dictionary classification and processing type can be performed simultaneously by pressing one operation button, thereby simplifying the button operation.

  According to a more preferred embodiment in the first aspect, a search is made for attribute data equal to the matching input data in a database in which attribute data is related to each other, and another attribute related to the searched attribute data. Attribute data output means for extracting and outputting data is further provided. Further, the attribute data extraction means prompts the user to select and input the output attribute data, and the output based on the match input data different from the match input data extracted by the match input data extraction means Further, attribute data different from the attribute data thus output is further output. Therefore, when the attribute data is searched for each attribute in the database in which the attribute data is associated with each other, and the target combination of attribute data is searched while sequentially narrowing down, the next input data candidate is displayed to the user. Thus, user input can be assisted.

  According to another preferred embodiment in the first aspect, the database is a database storing schedule data, and attribute data of “time”, “location”, and “schedule content” are related to each other. The “schedule content” is voice data. Therefore, when a user inputs “time” or “location” by voice, dictionary data corresponding to these attributes can be selected by button operation, so that the accuracy of voice recognition is improved and scheduling data can be easily input / output. It can be carried out. Furthermore, the “schedule contents” in the schedule data can be input / output by voice, and can be easily input / output even with a large amount of information compared to the conventional character input using buttons.

  According to another preferred embodiment in the first aspect, the database is a database of control commands for an external device, and the attribute data output means extracts the control commands for the selected external device. Is output. Therefore, the user can input image processing commands to an external printer or image display device by voice input and button operation, and can send a series of commands to the printer, etc. along with the image data. Use a mobile phone as an input device for external devices can do.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the technical scope of the invention is not limited to these embodiments, but extends to the matters described in the claims and equivalents thereof.

  FIG. 1 is a schematic configuration diagram of a system in the present embodiment. The mobile phone in this embodiment includes a voice recognition unit, dictionary data, and a schedule function in addition to a voice input / output unit for a voice communication function, a button input unit in which operation buttons are arranged, and an image display unit. ing. The operation buttons are arranged in a numeric keypad matrix, simplifying the input operation under such conditions. However, a portable information processing device such as a PDA may be used as long as it has an equivalent configuration, and operation button arrangements other than the numeric keypad format are also possible.

  A portable information processing apparatus 100 corresponding to a mobile phone includes a storage unit 10, a control unit 20, a button input unit 30, a voice input unit 40, a voice recognition unit 50, an information processing unit 60, and an output unit 70. Are connected by a signal line 101. The storage unit 10 stores a database 16 having dictionary data 12 and schedule data 14. Further, the dictionary data 12 stores dictionary data such as time information dictionary data 122 and location information dictionary data 124, and the database 16 stores time information attribute data 142, location information attribute data corresponding to the attributes of the schedule data 14. Data such as 144 and schedule content attribute data 146 are stored.

  The button input unit 30 receives the operation button pressing signal 32 and supplies it to the information processing unit 60. The information processing unit 60 determines the position of the pressed operation button based on the signal, and selects dictionary data of information to be referred to from the dictionary data 12, for example, time information dictionary data 122. Further, it determines the type of information processing based on the coincidence input data 62 that matches the recognition input data 52 recognized by the voice recognition unit 50, for example, the call of schedule data. On the other hand, the voice input unit 40 receives the voice data 42 spoken by the user and supplies it to the voice recognition unit 50. The voice recognition unit 50 recognizes the voice data and converts it into recognition input data 52 such as a character string. To the information processing unit 60.

  The information processing unit 60 refers to the information of the selected dictionary data 12 and extracts the matching input data 62 corresponding to the index data matching the recognition input data 52, or refers to the schedule data 14 and matches the matching input data. Data search and matching processing such as search for attribute data matching 62 is performed. The attribute data in the schedule data 14 is input / updated according to the type of information processing of the coincidence input data 62.

  The output unit 70 displays the extracted attribute data supplied from the information processing unit 60 on the screen of the mobile phone by the image display unit 80. If the attribute data of the schedule content 126 is audio data, the audio output unit 90 reproduces the audio.

  In another embodiment, when a mobile phone is used as an input device for an external device, the contents of the dictionary data 12 and the schedule data 14 of the database 16 are dictionary data for command variables for the external device, and commands for the variables. Replaced with the corresponding algorithm data. Based on the user input, the command extracted by the information processing unit 60 is transmitted to the cable-connected external image display device 200 via the output unit 70 together with the image data or the like, or the external image forming device 220 or the external file device Radio is transmitted from the antenna 72 to 240.

  FIG. 2 is a diagram illustrating the data structures of the dictionary data 12 and the schedule data 14 (database 16) and the relationship between the data. In this example, the time information dictionary data 122, the location information dictionary data 124, and the schedule data 14 of the dictionary data 12 are described in a table format for convenience. The dictionary data of the time information dictionary data 122 and the location information dictionary data 124 has index data in columns C122-1 and C124-1, respectively.

  These dictionary data are used to search the matching input data 62 corresponding to the index data that matches the recognition input data 52 of the user's speech. Which information dictionary data is to be referred to is specified by the user's button operation. That is, instead of searching for a wide range of dictionary data that matches the speech-recognized recognition input data 52, the attributes of the match input data 62 are classified in advance according to the information processing intended by the user. By specifying and referring to the dictionary data corresponding to the classification, it is possible to improve the efficiency of the search for matching input data. For example, by specifying within the range of the time information dictionary data 122 and the location information dictionary data 124 classified according to the information processing intended by the user such as schedule management, the efficiency of voice recognition can be improved.

  The schedule data 14 has attribute data such as time information attribute data 142, location information attribute data 144, schedule content attribute data 146, and attribute data corresponding to each attribute is stored in association with each other. These associated attribute data form schedule data having information on when, where, and what to do.

  Since the coincidence input data 62 in the dictionary data 12 has the same attributes (data type and data length) as the attribute data related in the schedule data 14, the coincidence input data 62 is used to input data to the schedule data 14. , Matching with the attribute data in the schedule data 14 can be performed.

  The time information dictionary data 122 will be described as an example. Assuming that the user speaks and the recognition input data 52 recognized by the voice recognition unit is a character string “Nano”, index data “Nano” that matches the recognition input data is stored in the time information dictionary data 122. Since the time information dictionary data “7 days” exists in R122-2, this time information dictionary data “7 days” corresponds to the index data corresponding to the recognition input data 52 “Nano”. This is input data 62. Then, there are a plurality of attribute data “7 days” corresponding to the time information attribute data 142 of the schedule data 14 as shown in the figure, and the time attribute data “PM 3 o'clock”, “AM 9 o'clock”, and “AM 12 o'clock” are stored. It is related.

  Similarly, when “PM3 o'clock” is selected from the time information attribute data at the above time and the user speaks, the matching input data 62 corresponding to the index data matching the recognition input data 52 “Sanji”. “PM3 o'clock” can be extracted from the row R122-1. Here, “PM3 o'clock” is coincident input data different from the previously extracted coincident input data “7 days”. Then, there are a plurality of attribute data “PM3 o'clock” related to the time information attribute data “7 days” of the schedule data 14 extracted earlier as shown in the figure. Here, “PM 3 o'clock” is extracted as attribute data different from the previously extracted time information attribute data “7 days”. The location information attribute data “city hall”, “school”, and “home” are further related to “7 days” and “PM 3 o'clock”.

  Next, the matching input data 62 “city hall” in the location information dictionary data 124 that matches the recognition input data 52 “Shiyakusho” based on the user's utterance can be extracted (line R124). Then, in the place information attribute data “city hall” related to the time information attribute data “7 days” and “PM 3 o'clock” in the schedule data 14, the schedule content attribute data “acquire resident's card” and “acquire family register copy” "Is further related.

  As described above, the match input data 62 extracted from the time information dictionary data 122 and the location information dictionary data 124 correspond to the time information attribute data 142 and the location information attribute data 144 in the schedule data 14, respectively. Are associated with schedule content attribute data 146. Therefore, in order to call up the desired schedule content attribute data, the time information attribute data and the location information attribute data may be narrowed down sequentially.

  Note that neither the dictionary data 12 nor the schedule data 14 (database 16) need have a table structure as shown. In the dictionary data 12, it is only necessary that index data matching the recognition input data 52 and matching input data 62 corresponding thereto are stored in association with each other. The schedule data 14 may also be associated with attribute data.

  FIG. 3 is a flowchart for explaining the entire schedule information calling process according to the present embodiment. 4 to 9, an information input procedure in a mobile phone having a schedule function will be described with reference to FIG.

  FIG. 4 is a diagram for explaining the correspondence relationship between the arrangement of the operation buttons and the classification of the dictionary data 12, the schedule data 14, and the information processing type determination of the matching input data 62. The button input unit 30 has three operation buttons “1” to “9” arranged in rows and columns in the vertical and horizontal directions. Based on the position of the operation button to be pressed, the attribute of the match input data 62 is determined. Further, the type of dictionary data 12 corresponding to the attribute and the type of information processing based on the matching input data 62 are determined. Specifically, for each of the operation buttons, the horizontal rows BR1 to BR3 indicate the classification and schedule data attributes of the dictionary data 12 to be selected, and the vertical columns BC1 to BC3 indicate information processing based on the matching input data 62. The types TP are associated with each other. Therefore, the two-dimensional options of the classification of the dictionary data 12 to be selected and the information processing type based on the matching input data 62 are assigned to one button.

  For example, the buttons “1”, “2”, “3” correspond to the row BR1 of buttons for selecting the time information dictionary data 122, and the buttons “3”, “6”, “9” correspond to the matching input data 62. This corresponds to the column BC3 that performs the process “calling” based on. Therefore, a function of “calling” the matching input data 62 with reference to the time information dictionary data 122 is assigned to the button “3”.

  Specifically, when the user speaks while pressing the button “3”, the information processing unit 60 that receives the input from the button input unit 30 refers to the dictionary to be referred to based on the position of the pressed button. It is determined that the classification of the data 12 is the time information dictionary data 122 (step S10). On the other hand, the voice recognition unit 50 recognizes the voice data 42 supplied from the voice input unit 40 and converts it into recognition input data 52 (step S20). Further, the information processing unit 60 determines that the type of processing based on the matching input data 62 that matches the recognition input data 52 is “calling”.

  Similarly, the row BR2 corresponds to the location information dictionary data 124, the row BR3 corresponds to the schedule content attribute data 146, the column BC1 corresponds to the process “growth” based on the recognition input data 52, and the column BC2 corresponds to the process “input”. is doing. The process “call” refers to a process of acquiring and outputting the existing schedule information from the schedule data 14 based on the match input data 62, and the process “input” refers to a new process based on the match input data 62. The process of registering schedule information in the schedule data 14 is referred to as the process “nurturing” refers to a process of performing new registration that can be extracted into the dictionary data 12 based on the recognition input data 52.

  As described above, the information processing corresponding to the attribute of the matched input data refers to searching for the attribute data matching the matched input data having the “time” attribute from the schedule data 14 or the matching having the “location” attribute. This means that different information processing is performed according to different attributes of the matched input data, such as searching for attribute data that matches the input data. Furthermore, the type of information processing refers to searching / extracting (calling) or inputting (registering) attribute data matching the matched input data in each of these different information processings. In this way, the user can select a dictionary according to the purpose of information processing by associating dictionary data classifications with two-dimensionally arranged operation button rows and information processing types based on matched input data with operation button columns. Can be done intuitively. In addition, selection of different two-dimensional options such as dictionary classification and processing type can be performed simultaneously by pressing one operation button, so that the button operation is simplified. Furthermore, since the utterance is performed simultaneously with the button operation, the dictionary classification, processing type, utterance content and three-dimensional different inputs can be given to the mobile phone at the same time, and simple operability can be realized. Note that the operation buttons do not necessarily have to be arranged in a matrix as long as a plurality of options are assigned to one operation button.

  A screen transition example displayed by the image display unit 80 when the process “call” is performed by operating the buttons “3”, “6”, and “9” positioned in the column BC3 of the button input unit 30 as described above. 5 to 7 will be used for explanation.

  FIG. 5 is a diagram for explaining an input waiting screen when the schedule function of the mobile phone displayed by the image display unit 80 is called. By displaying a calendar as an initial screen and displaying “when?” On the calendar, the user is prompted to input regarding the time information attribute data 142.

  FIG. 6 is a diagram illustrating the input operation on the instruction waiting screen in FIG. 5, the correspondence between data, and a display screen example in association with each other. First, FIG. 6A illustrates a case where a date is input for the time information dictionary data 122. When the user speaks “What?” While pressing the button “3”, the push button pressing signal 32 is supplied to the button input unit 30 and the voice data 42 is supplied to the voice input unit 40 (steps IN32 and IN42 in FIG. 3). ). In response to the push signal 32 of the button “3”, the information processing unit 60 selects the time information dictionary data 122 (step S10) and matches the recognition input data 52 “Nano” recognized by the voice recognition unit 50 (step S20). The matching input data 62 corresponding to the index data to be extracted is extracted from the dictionary (step S30). Note that the steps after step S10 and step S20 may be reversed or simultaneous.

  As described with reference to FIG. 2, since the coincidence input data 62 “7 days” exists in the time information dictionary data 122, the information processing unit 60 extracts the time information dictionary data “7 days”. When the time information attribute data 142 of the schedule data 14 is searched (step S40 in FIG. 3), there are a plurality of attribute data “seven days”, each of which includes time information attribute data, location information attribute data, Schedule content attribute data is related.

  Therefore, as shown in FIG. 6A, the image display unit 80 displays a list of times and places related to “7 days” (step S42), and further displays “What time?” And is selected by the user. Prompt for input. Therefore, when inputting time to narrow down the schedule data, the user may speak the time selected from the list while pressing the operation button again (YES in step S50).

  Next, FIG. 6B illustrates a case where time is input for the time information dictionary data 122. As described above, the user speaks “Sanji” while pressing the button “3”. In response to the push signal 32 for the button “3”, the information processing section 60 selects the time information dictionary data 122 (step S10 in FIG. 3), and the speech recognition section 50 recognizes (step S20). The matching input data 62 corresponding to the index data matching "" is extracted from the dictionary (step S30). Then, time information “PM3 o'clock” is extracted.

  Then, when the time information attribute data 142 of the schedule data 14 is searched (step S40), a plurality of attribute data “PM3 o'clock” exists as shown in FIG. Content attribute data 146 is associated. Therefore, as shown in FIG. 6 (B), the location related to “3 o'clock” of “7 days” is displayed on the image display unit 80, and further “Where?” Is displayed and the user selects the location. The input is prompted (step S42 in FIG. 3). Therefore, the user only has to speak the place selected from the list while pressing the operation button again when inputting the place (YES in step S50).

  FIG. 7 is a diagram illustrating the location information input operation, data correspondence, and a display screen example in association with each other. The user speaks “Yakusoku” while pressing the button “6”. In response to the push signal 32 of the button “6”, the information processing unit 60 selects the location information dictionary data 124 (step S10 in FIG. 3), and the speech recognition unit 50 recognizes (step S20). The matching input data 62 corresponding to the index data matching "" is extracted from the dictionary (step S30). Then, the coincidence input data 62 “city hall” is extracted.

  Then, when the location information 144 of the schedule data 14 is searched (step S40), there are a plurality of attribute data “city halls” as shown in FIG. Acquired certified copy of family register ”. Therefore, as shown in FIG. 7, “What?” Is displayed on the image display unit 80 so that the user inputs words related to the schedule contents related to “7th” and “City Hall”, and the user inputs a selection. (Step S42 in FIG. 3). Further, the existence of the schedule content attribute data 146 related to the “city hall” of the location information 144 can be expressed by the graphic P1 of FIG.

  FIG. 8 is a diagram for explaining the input operation, the data correspondence, and the output in a case where the input of the schedule content is prompted. The user performs input for outputting the schedule content attribute data 146 already extracted (NO in step S50, S60). Therefore, the user speaks “Juminhyo” while pressing the button “9”. In response to the press signal 32 of the button “9”, the information processing section 60 determines that the speech regarding the schedule contents has been made and the voice data 42 has been input.

  In this embodiment, since the schedule content attribute data 146 is stored as audio data, the information processing unit 60 compares the audio data 42 with the schedule content attribute data 146 that is the stored audio data, and schedule The content attribute data 146 is extracted in the order approximate to the audio data 42 and is output from the audio output unit 90. That is, since the voice data 42 most closely resembles “Juminhyo” is the schedule content attribute data “acquisition of resident's card”, the voice “Juminhyo no Shutoku” is reproduced (step S60). . Alternatively, the voice data can be reproduced by selecting the figure P1 (FIG. 7) expressing the existence of the schedule content attribute data 146 by operating the button without speaking.

  According to the input procedure described above, the user can select dictionary data corresponding to these attributes by an intuitive button operation when inputting “time” or “location” by voice. Compared with the case of inputting K, the operation can be performed easily. In addition, since dictionary data classified by attribute of coincident input data is referred to, the accuracy of speech recognition is improved, and input / output to schedule data can be performed efficiently. Furthermore, the desired “schedule contents” in the schedule data can be input / output by voice, and can be easily input / output even with a large amount of information compared to character input using conventional buttons.

  5 to 8 show a case where schedule data is called in the order of time, place, and schedule content, a modified example of the above input example will be described with reference to FIG.

  FIG. 9 is a diagram for explaining an example of a display screen when inputting from location information. First, when the user activates the schedule function of the mobile phone, a calendar as shown in FIG. 9A is displayed. Then, as shown in FIG. 9B, which is an enlarged view of a part of the calendar, for example, the time information “14th” may contain schedule information related to the location information “school” and “city hall”. It is displayed using the figure P1. That is, the time information is a screen example that displays a list without narrowing down and displays location information and schedule information related to the time information.

  Here, when the user utters and inputs, for example, the location information “City Hall” while pressing the operation button in accordance with the input procedure of the location information 142 described above, the audio of the schedule content related to “City Hall” is reproduced. be able to.

  Further, a map can be displayed as shown in FIG. 9C when the schedule function is activated. In this case, the date when the user activates the schedule function is used as the time information. FIG. 9D, which is an enlarged view of the same figure, shows that the schedule contents related to the place information “city hall” are stored at the location of the city hall on the map using the graphic P1. The Here, when the user utters and inputs, for example, the location information “City Hall” while pressing the operation button in accordance with the input procedure of the location information 142 described above, the audio of the schedule content related to “City Hall” is reproduced. be able to.

  According to the modified example as described above, it is possible to confirm the schedule contents by a more shortened procedure using an intuitive display form such as a calendar or a map.

  10 to 14 illustrate a procedure when the type of processing based on the coincidence input data 62 is “input”.

  FIG. 10 is a flowchart for explaining the overall flow of the input procedure. A procedure following step S30 after step S10 to step S30 in FIG. 3 is described. Screen transition of the input procedure of FIGS. 11 to 13 will be described with reference to the flowchart.

  FIG. 11 is a diagram illustrating the input operation, the data correspondence, and the display screen example in association with each other when inputting the date for the time information attribute data 142. The user speaks “Yoka” while pressing the button “2”. In response to the pressing signal 32 of the button “2”, the information processing unit 60 selects the time information dictionary data 122 (step S10 in FIG. 3), and the speech recognition unit 50 recognizes (step S20). The matching input data 62 “8 days” corresponding to the index data matching “” is extracted from the dictionary (step S30).

  The information processing section 60 stores the coincidence input data 62 “8 days” in the schedule data 14 because the column in which the button “2” is located is the column BC2 (FIG. 4) corresponding to the process “input”. Is executed (step S140 in FIG. 10). As a result of such processing, the image display unit 80 displays that the time information attribute data “8 days” is stored as shown in FIG. 11, and further displays “Where?” To the user and “8 days”. Prompt for the location associated with the.

 FIG. 12 is a diagram illustrating the input operation, the data correspondence, and the display screen example when inputting location information for the location information attribute data 144. The user speaks “Gakukko” while pressing the button “5”. In response to the pressing signal 32 of the button “5”, the information processing section 60 selects the location information dictionary data 124 (step S10 in FIG. 3), and the speech recognition section 50 recognizes (step S20). The matching input data 62 “school” corresponding to the index data matching “Kou” is extracted from the dictionary (step S30).

  The information processing unit 60 performs processing for registering “school” in the schedule data 14 as new attribute data for the location information attribute data 144 in association with the new attribute data “8 days” of the time information attribute data 142. (Step S140 in FIG. 10). As a result of such processing, the image display unit 80 displays that the location information attribute data “school” is registered after the time information attribute data “8 days”, and further displays “What?”. The user is prompted to input schedule contents related to “8 days” and “school”.

  FIG. 13 is a diagram illustrating the schedule content attribute data 146 in association with input operations, data correspondences, and display screen examples in association with each other. The user speaks “Juyosankan” while pressing the button “8”. In response to the press signal 32 of the button “8”, the information processing unit 60 determines that the attribute of the input voice data corresponds to the schedule content attribute data 146 in the schedule data 14, and the already associated time information attribute In association with the data “8 days” and the place information attribute data “school”, the voice data “Juyosankan (visiting class)” is registered in the schedule data 14 as the schedule content attribute data 146 (step S140 in FIG. 10).

  As a result of such processing, as shown in FIG. 13, the image display unit 80 is associated with the time information attribute data “8 days” and the location information attribute data “school” in the schedule content attribute data 146 as “class visit”. "Is registered as voice data in FIG. P1. Then, when the input of attribute data necessary as schedule data is completed, the process ends (NO in step S150 of FIG. 10).

  Through the above procedure, the user can intuitively select a dictionary according to the attribute of the information to be input, so that the button operation is simplified and the speech recognition of the spoken speech data can be performed with high accuracy. , Schedule data can be registered efficiently.

  FIG. 14 is a diagram illustrating the input operation when the type of information processing based on the recognition input data 52 is “nurturing”, the correspondence relationship between the data, and a display example of the screen. For example, when the user performs a “calling” or “input” process, the matching input data 62 corresponding to the index data matching the recognition input data 52 corresponding to the speech “Hatsuka” is the time information dictionary data 122. Is not present. In such a case, the dictionary can be trained by inputting time information that matches the utterance “Hatsuka”.

  When the user speaks “Hatsuka” while pressing the button “1”, the information processing unit 60 receives the pressing signal 32 of the button “1” and the attribute of the input voice data corresponds to the time information dictionary data 122 It is determined that the training process for the recognition input data 52 is performed. Therefore, the image display unit 80 displays a calendar, and if the user specifies “20 days” by a button operation, the time information “20 days” corresponds to the recognition input data 52 “Hatsuka”. It can be registered in the information dictionary data 122. Therefore, the user can improve the accuracy of subsequent speech recognition by appropriately increasing the dictionary data as candidates for the matching input data 62.

  In FIGS. 15 to 19, when a mobile phone is used as an input device for an external device, a processing command for the image data is transmitted to the external device together with the image data stored in the storage unit 10 of the mobile phone. Explains the form. As described with reference to FIG. 1, in this embodiment, the contents of the dictionary data 12 are dictionary data of variables that specify commands for external devices, and the contents of the database 16 are algorithm data in which commands are associated with the variables. Replace.

  FIG. 15 is a diagram for explaining the correspondence relationship between the button input unit 30, the classification of the dictionary data 12, and the information processing type based on the coincidence input data 62. Based on the position of the operation button to be pressed, the attribute of the matching input data 62 is determined, and the type of the dictionary data 12 corresponding to the attribute and the type of information processing based on the matching input data 62 are determined. That is, the buttons “1”, “2”, and “3” correspond to the row BR11 of the button for selecting the subject information dictionary data 123 of the dictionary data 12, and the row BR21 corresponds to the color processing information dictionary data 125. The row BR31 corresponds to the number information dictionary data 127 at the time of printing. For example, the contents of the subject information dictionary data 123 are variables such as “person”, “sports”, “indoor”, and “landscape”.

  On the other hand, the vertical column BC11 is used when the external device used is an external file device 240 such as a PC, BC21 is used when the external device used is the external video display device 200 such as a television device, and BC31 is used. This corresponds to the processing type TP when the external device is the external image forming apparatus 220 such as a printer.

    Note that the operation buttons do not necessarily have to be arranged in a matrix as long as a plurality of options are assigned to one operation button.

  FIG. 16 is a diagram for explaining algorithm data in which a command is associated with an input variable. FIG. 16A is a table format of the algorithm data 15A in which the variable of the subject information dictionary data 123 is associated with the column 123a and the image processing command corresponding to the variable is associated with the column 123b. For example, when the subject “person” is input, a command for performing image processing for “recognizing the face and adjusting the skin color to a predetermined color” is extracted (line R123). FIG. 16B represents the algorithm data 15B in the form of a table in which each variable of the color processing information dictionary data 125 is associated with the column 125a and the color processing command corresponding to the variable is associated with the column 125b. For example, when color processing “bright” is selected, a command to be subjected to color processing for “correcting the brightness high” is extracted (line R125). These algorithm data are expressed in a table format for convenience, but may not be in a table format as long as variables and commands are associated with each other.

  FIG. 17 is a diagram for explaining the algorithm data 16 in which the type of processing corresponding to the external device to be used is related to the column 129a and the image processing command corresponding to the type is related to the column 129b. For convenience, the table format is used. However, the table format may not be used as long as the external device type and the image processing command are associated with each other. For example, when image data is transferred to an external file device 240 such as a PC, a command for “image processing so as to increase the compression rate” is extracted (line T11).

  Screen transitions taking as an example a processing procedure for transferring image data to the external image forming apparatus 220 will be described with reference to FIGS. Further, FIG. 3 is applied to the flowchart of the processing procedure.

  FIG. 18 is a diagram illustrating an example of an initial screen displayed on the image display device. First, in the display screen example of FIG. 18A, “Subject?” Or the like is displayed to prompt the user to input. FIG. 18B illustrates the subject information input operation, data correspondence, and display screen example in association with each other. The user utters “Jimbutsu” while pressing the button “3”. Upon receipt of the button “3” pressing signal 32, the information processing section 60 selects the subject information dictionary data 123 (step S10 in FIG. 3), and the speech recognition section 50 recognizes (step S20). The matching input data 62 “person” corresponding to the index data matching “” is extracted from the dictionary (step S30).

  Then, the attribute data “person” that matches the match input data 62 “person” is searched from the subject information 123 of the algorithm data 15A of FIG. 16A (step S40 in FIG. 3), and is associated with the attribute data. Extract image processing commands. Further, the information processing section 60 determines from the column of the button “3” that the target device to be used is the external image forming apparatus 220, and performs the corresponding “image processing for printing” in the algorithm data 16 of FIG. Extract commands. Further, as shown in FIG. 18B, the user is prompted to input by displaying “color processing?” Or the like on the image display device 80.

  FIG. 19 is a diagram illustrating the color processing information input operation, data correspondence, and a display screen example in association with each other. The user speaks “Akaruku” while pressing the button “6”. In response to the pressing signal 32 of the button “6”, the information processing section 60 selects the color processing information dictionary data 125 (step S10 in FIG. 3), and the speech recognition section 50 recognizes (step S20). The matching input data 62 “bright” corresponding to the index data matching “” is extracted from the dictionary (step S30). Then, the attribute data “bright” is searched from the color processing information 125a of the algorithm data 15B (step S40), and the color processing command “correct lightness” associated with the attribute data is extracted. Further, as shown in FIG. 19, “how many?” Is displayed on the image display device 80 to prompt the user to input.

  FIG. 20 is a diagram for explaining the number information input operation, data correspondence, and output in association with each other. The user utters “Sweet” while pressing the button “9”. In response to the pressing signal 32 of the button “9”, the information processing section 60 selects the number information dictionary data 127 (step S10 in FIG. 3), and the speech recognition section 50 recognizes (step S20). The matching input data 62 “5 sheets” corresponding to the index data matching “” is extracted from the dictionary (step S 30), and the extracted image processing command for the subject “person” and the color processing method “bright” are extracted. Are transmitted to the external image forming apparatus 220 together with the image data as a series of commands together with the color processing command and the image processing command for the printing press.

  According to the above procedure, the user can input image processing commands to an external printer or image display device by voice input and button operation, and can send a series of commands to the printer together with the image data. It can be used as an input device.

  As described above, in an information processing apparatus that obtains output based on input information of a plurality of different dimensions, the accuracy of voice recognition is simplified while simplifying the operation by combining operation buttons assigned with multi-dimensional options and voice input. And input can be performed efficiently.

It is a schematic block diagram of the system in this Embodiment. It is a figure explaining the data structure of each of the dictionary data 12 and the schedule data 14 (database 16), and the relationship between data. It is a flowchart figure explaining the whole calling process of the schedule information in this Embodiment. It is a figure explaining the correspondence of the information processing classification judgment of the arrangement | sequence of an operation button, the classification | category of the dictionary data 12, the schedule data 14, and the coincidence input data 62. FIG. It is a figure explaining the input waiting screen at the time of calling the schedule function of a mobile telephone which the image display part 80 displays. FIG. 6 is a diagram illustrating an input operation on the instruction waiting screen in FIG. 5, a correspondence relationship between data, and a display screen example in association with each other. It is a figure explaining in association with an input operation of place information, data correspondence, and a display screen example. FIG. 5 is a diagram illustrating an input operation, a data correspondence relationship, and an output when an input of a schedule content is prompted. It is a figure explaining the example of a display screen in the case of inputting from place information. It is a flowchart figure explaining the whole flow of an input procedure. FIG. 6 is a diagram illustrating an input operation when inputting a date for time information attribute data 142, data correspondence, and a display screen example in association with each other. It is a figure explaining in association with the input operation, the data correspondence, and the display screen example when inputting the location information for the location information attribute data 144. It is a figure explaining in association with an input operation, data correspondence, and a display screen example when inputting voice for schedule content attribute data. It is a figure explaining in association with the input operation in the case where the type of processing based on the recognition input data 52 is “nurturing”, the correspondence between the data, and the display example of the screen. 6 is a diagram for explaining a correspondence relationship between a button input unit 30, classification of dictionary data 12, and processing types based on matching input data 62. FIG. It is a figure explaining the algorithm data which matched the command with the variable to input. It is a figure explaining the algorithm data 16 which linked | related the kind of process type corresponding to the external apparatus to be used to the column 127a, and the image process command corresponding to the said type to the column 127b. It is a figure explaining the example of a screen of the initial screen displayed on an image display device. It is a figure explaining in association with input operation of color processing information, data correspondence, and a display screen example. FIG. 5 is a diagram illustrating an association between an input operation of sheet number information, data correspondence, and output.

Explanation of symbols

12: Dictionary data 122: Time information dictionary data 14: Schedule data 142: Time information attribute data 30: Button input unit 15A: Algorithm data

Claims (9)

In response to pressing of the operation button and voice input, matching input data matching the recognition input data recognized by the voice recognition device is extracted from the dictionary data, and the matching input data is extracted based on the matching input data. In a portable information processing device that performs information processing corresponding to attributes,
The information processing has a plurality of information processing corresponding to the attribute of the coincidence input data,
Based on the position of the pressed operation button among a plurality of operation buttons, the attribute of the match input data is determined, and the match input data matching the recognition input data is determined from the dictionary data corresponding to the determined attribute. A portable information processing apparatus having matching input data extraction means for extracting. In claim 1,
The plurality of operation buttons are arranged in a matrix,
The coincidence input data extracting means judges the attribute of the coincidence input data and the type of information processing based on the coincidence input data based on the row or column where the pressed button is located. . In claim 2,
A database of attribute data associated with each other;
If the determined type of information processing is output, search the attribute data equal to the matched input data in the database, and extract and output another attribute data related to the searched attribute data A portable information processing apparatus further comprising output means. In claim 3,
The attribute data output means prompts the user to select and input the output attribute data, and is output based on match input data different from the match input data extracted by the match input data extraction means. A portable information processing apparatus that further outputs attribute data different from the attribute data. In claim 3,
The database is a database of schedule data, and the attribute data is “time”, “location”, and “schedule content” which is voice data,
The attribute information output means reproduces the audio data when the extracted attribute data is "schedule content". In claim 5,
The portable information processing apparatus, wherein the attribute data output means reproduces “schedule content” which is voice data similar to the input voice. In claim 3,
The attribute data of the database is a control command for an external device,
The determined type of information processing is the type of the external device,
The portable information processing apparatus, wherein the attribute data output means outputs the extracted control command to the determined external device. In claim 2,
A database of attribute data associated with each other;
If the determined type of information processing is an input, it further comprises attribute data input means for associating the attribute of the match input data with the attribute of the database and inputting the match input data to the database,
The attribute data input means is a portable information processing apparatus comprising means for inputting the input voice into the database. In response to pressing of the operation button and voice input, matched input data that matches the input voice with the recognized input data recognized by the voice recognition device is extracted from the dictionary data. In the input / output method of the portable information processing apparatus for performing a plurality of information processing corresponding to the attribute of the matched input data with respect to the database related to
Based on the row or column where the pressed operation button is located among the operation buttons arranged in a matrix, the attribute of the matched input data and the type of information processing based on the matched input data are determined, and the determined A matching input data extraction step of extracting matching input data that matches the recognition input data from the dictionary data corresponding to the attribute;
If the determined type of information processing is output, the attribute data equal to the matched input data is searched in the database, and attribute data of another attribute related to the searched attribute data is extracted and output And an input / output method for a portable information processing apparatus.

Images