JP2005202552A - Sentence generation device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sentence generation device and method, capable of easily generating a sentence having expression corresponding to various characters even with the same contents. <P>SOLUTION: This sentence generation device has: a first storage means storing basic sentence data including basic contents of the sentence and a sentence end mark showing a sentence end position of the sentence; a second storage means storing a plurality of pieces of sentence end data showing sentence ends corresponding to the characters different from each other; a reading means selectively reading one piece of sentence end data among the plurality of pieces of sentence end data from the second storage means; and a sentence data production means inserting one piece of sentence end data into the sentence end position shown by the sentence end mark of the basic sentence data stored in the first storage means to produce sentence data. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a sentence generation apparatus and method for generating sentence data.

  2. Description of the Related Art An interactive device that performs a dialogue with a user by displaying a sentence on a display corresponding to the user's selection operation is known. In such an interactive device, various text data are stored in advance, and appropriate text data is selectively read out by a user's selection operation, and text corresponding to the read text data is displayed on the display. Is done.

  However, any text displayed on the display is expressed in a uniform manner, and if the content is the same, the text of the same expression is always output.

  The problem to be solved by the present invention includes the above-mentioned drawbacks as an example, and provides a sentence generation device and method capable of easily generating sentences of expressions corresponding to various characters even with the same content It is an object of the present invention.

  A sentence generation device according to a first aspect of the present invention is an apparatus for generating sentence data, wherein the first storage means stores basic sentence data including basic contents of a sentence and a ending mark indicating the ending position of the sentence; Second storage means storing a plurality of ending data indicating endings corresponding to different characters, and reading means for selectively reading out one ending data of the plurality of ending data from the second storage means; Document data creating means for creating the sentence data by inserting the first ending data at the ending position indicated by the ending mark of the basic sentence data stored in the first storage means. It is said.

  The sentence generation method of the invention according to claim 8 is a method for generating sentence data, and stores basic sentence data including basic contents of a sentence and a ending mark indicating the ending position of the sentence in the first storage means. A first storage step, a second storage step of storing a plurality of ending data indicating endings corresponding to different personalities in the second storage means, and selectively reading out one ending data of the plurality of ending data A reading step; and a document data creation step for creating the sentence data by inserting the ending data of the first sentence at the ending position indicated by the ending mark of the basic sentence data stored in the first storage means. It is characterized by that.

  A program according to claim 9 is a computer readable program for executing a sentence generation method for generating sentence data, and includes basic sentence data including a basic content of a sentence and a ending mark indicating the ending position of the sentence. In the first storage means, a second storage step in which a plurality of ending data indicating endings corresponding to different personalities are stored in the second storage means, and one of the plurality of ending data A step of selectively reading the ending data of the first sentence, and creating the sentence data by inserting the ending data of the first sentence at the ending position indicated by the ending mark of the basic sentence data stored in the first storage means. And a document data creation step.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows an interactive apparatus to which a sentence generating apparatus of the present invention is applied. The interactive apparatus includes an operation unit 1, a display unit 2, a sound generation unit 3, a controller 4, a RAM 5, a ROM 6, and a speaker 7. Each of the operation unit 1, the display unit 2, the sound generation unit 3, the RAM 5, and the ROM 6 is connected to the controller 4.

  The operation unit 1 includes a keyboard or a series of switches, and is a part where a user inputs information such as personal information or commands an operation. The display unit 2 displays talk contents and instructions to the user in accordance with an instruction from the controller 4. The voice generation unit 3 generates the content of talking to the user as a synthesized voice signal in response to a command from the controller 4. The output audio signal of the audio generator 3 is supplied to the speaker 7.

  The RAM 5 stores user personal information by an initial setting operation (not shown). The personal information includes at least the user's character and formal index. The character is an interactive device for the user, that is, a human character such as a speaker's personality, taste, age, gender and the like. In this embodiment, the characters are divided into adult female, adult male, female, male, and elderly characters. However, the present invention is not limited to this, and it may be classified according to personality or occupation such as a gentle person or a tempered person. Furthermore, it is not limited to human beings, and other animals and plants may be used. The formal index is a numerical value that determines a formal style to be spoken to the user of the interactive apparatus together with a formal degree (formal degree) described later, and is set in the range of 1 to 9, for example. For example, the higher the numerical value of each of the formal index and the formal degree, the harder the spoken form, and the lower the value, the more formal the index.

  The RAM 5 also has a usage count storage area for storing the usage counts of the formal degrees 1 to 9. Since the formal degree is set at random according to the formal index, the number of uses of the formal degree corresponding to each setting of the formal degree is increased by one.

  The ROM 6 stores an operation program for the controller 4, a basic sentence data group, and a ending data table. The basic sentence data indicates basic sentences used for dialogue with the user. The basic sentence data has a style such as a positive sentence, a question sentence, and a negative sentence. Further, as will be described later, a ending mark indicating the position of the ending and the type of style is added to the basic sentence data.

  As shown in FIG. 2, the ending data table shows ending data indicating the ending of an affirmative sentence for each formal degree for each of the adult female, adult male, female, male, and elderly who are character types. Of course, not only affirmative sentences but also ending data tables are formed for other sentence styles such as question sentences and negative sentences. That is, a ending data table is provided for each type of ending mark described later.

  The controller 4 is composed of a microcomputer and controls the entire interactive device. When the user performs a predetermined operation on the operation unit 1, for example, an operation for starting a dialogue or a selection device, a signal indicating the operation content is supplied from the operation unit 1 to the controller 4.

  When the controller 4 generates a sentence for the user, as shown in FIG. 3, the controller 4 reads out basic sentence data corresponding to the item selected by the user's operation in the operation unit 1 from the ROM 6 (step S1). A ending mark is detected in the basic sentence data (step S2). The basic sentence data is composed of a text file, for example, and ending marks are added to the basic sentence data whose endings can be changed. For example, when the basic sentence data indicates an affirmative sentence, # is added as an ending mark, and when the basic sentence data indicates a question sentence? Is added as a ending mark. The position of the ending mark is the position where the ending data is added in step S8 described later.

  The controller 4 reads the initially set formal index as X from the RAM 5 (step S3). The formal index is initialized to any one value from 1 to 9 and stored in the RAM 5. The formal index X is applied to the random function f (X, RND) to calculate Y = f (X, RND) (step S4). Y calculated using the random function f (X, RND) is the formal degree. The random function f (X, RND) generates an integer value Y in the range of 1 ≦ Y ≦ 9 according to the formal index X. By using this random function, the appearance probability of each value that Y can take for the formal index X = 1 to 9 can be set as shown in FIG. The appearance probabilities in FIG. 4 are set as higher probabilities as the formal degree is higher in the past when X = 4. Further, it may be an appearance probability according to a commonly used ending regardless of the number of times of use. Further, the appearance probability may always be constant, or may change with time.

  The controller 4 further reads the initially set character from the RAM 5 (step S5). Further, the number of uses in the use count storage area of the RAM 5 is increased by 1 corresponding to the formal degree Y (step S6). If the number of uses of each formal degree is used, it can be reflected in the calculation of Y = f (X, RND) so as to have a set appearance probability.

  The controller 4 reads the ending mark in the detected basic sentence data, the calculated formal degree Y, and the ending data corresponding to the read character from the ending data table of the ROM 6 (step S7). Specifically, first, the ending data table corresponding to the ending mark detected in step S2 is selected from the ROM 6, and the formal degree Y determined in step S4 and the character read in step S5 in the selected ending data table are selected. The corresponding ending data is read from the selected ending data table. When the ending data is read, output sentence data is created by replacing the ending mark portion in the basic sentence data with the ending data (step S8).

  When the controller 4 creates the output sentence data, the controller 4 supplies the output sentence data to the display unit 2 to display the sentence indicated by the output sentence data (step S9).

  It is assumed that the basic sentence data read in step S1 indicates “go to the park together #”. # Is a ending mark indicating an affirmative sentence. If the initial character is an adult female and the formal degree Y is set to 1 in step S4, “yo” is read from the ending data table of the ROM 6 and “all go to the park”. Output text data is created. As a result, the display unit 2 displays “Let's go to the park together.”

  FIG. 5 shows an example of a sentence indicated by the original sentence, basic sentence data, and sentence data in the case of an affirmative sentence in relation to the character and the formal degree.

  It is assumed that the basic sentence data read in step S1 indicates “what do you eat in the morning?”. ? Is a ending mark indicating a question sentence. If the initially set character is an elderly person and the formal level Y is set to 5 in step S4, “Nanji” is read from the ending data table of the ROM 6, and “What do you eat in the morning?” Output text data is created. As a result, the display unit 2 displays "What do you eat in the morning?"

  FIG. 6 shows an example of a sentence indicated by the original sentence, basic sentence data, and sentence data in the case of a question sentence in relation to the character and the formal degree.

  In the embodiment described above, since the formal degree Y is changed based on the initial formal index X, it is possible to generate a sentence with a ending expression corresponding to a slightly changed character.

  In the embodiment described above, the text is displayed on the display unit 2 according to the output text data. However, the controller 4 may supply the output text data to the voice generation unit 3. When the output sentence data is supplied to the voice generation unit 3, the voice generation unit 3 generates a synthesized voice signal corresponding to the output sentence data, and the speaker 7 is driven according to the synthesized voice signal. A sound corresponding to the output text data is output from the speaker 7.

  In the case of an affirmative sentence, the ending mark in the basic sentence data may be formed by arranging two punctuation marks at the end of the sentence (end of sentence) like “..”. In the case of a question sentence, it may be formed by arranging two question marks at the end of the sentence, such as “??”. FIG. 7 shows output sentence data to which the original sentence, basic sentence data, and ending data are added in the case of the affirmative sentence in FIG. 5 using “..” as the ending mark and the question sentence in FIG. 6 using “??”. The example of the sentence shown by is shown in relation to the character and the formal degree. In this way, if a sequence of the same symbols that cannot exist in a normal sentence is used as the ending mark, the restriction on the symbol used for the ending mark is reduced.

  In the embodiment described above, the ending data determined from the formal degree Y and the character in step S7 is single, but a plurality of ending data determined from the formal degree Y and the character may be prepared. . This may not be possible with the same ending depending on the sentence content of the basic sentence data, so a plurality of ending data are prepared and can be selected. For example, in the case of an affirmative sentence, the ending point of the sentence is usually “..” as an ending mark, and if the sentence is an affirmative sentence that cannot be handled by the same ending, the ending mark is “. "And" ... ", different ending data are selected even for the same formal degree Y and character.

  Furthermore, in the embodiment described above, the position of the ending mark in the basic sentence data is the position to insert the ending data as it is, but the ending mark in the basic sentence data may only indicate the position of the ending of the sentence. . That is, if the ending mark indicates the position of the ending of the sentence, it may be placed at the beginning of the basic sentence data.

  As described above, according to the present invention, the first storage means that stores the basic sentence data including the basic content of the sentence and the ending mark indicating the ending position of the sentence, and the plurality of endings corresponding to different characters. Second storage means for storing the ending data, reading means for selectively reading one ending data of the plurality of ending data from the second storage means, and ending of the basic sentence data stored in the first storage means Document data creating means for creating sentence data by inserting one ending data at the ending position indicated by the mark, so that sentences of ending expressions corresponding to various characters can be easily obtained even with the same contents Can be generated.

  Furthermore, the sentence generation device of the present invention can be applied to an alarm clock, a communication device for interactive communication with a user, a mobile phone with an alarm function, a PDA, game software, and the like.

It is a block diagram which shows the Example of this invention. It is a figure which shows the ending data table for positive sentences. It is a flowchart which shows operation | movement of the controller in the apparatus of FIG. It is a figure which shows the relationship between the past use frequency and appearance probability. It is a figure which shows the example of the sentence which the original sentence in the case of a positive sentence, basic sentence data, and sentence data show in relation to a character and a formal degree. It is a figure which shows the example of the sentence which the original sentence in the case of a question sentence, basic sentence data, and sentence data show in relation to a character and a formal degree. It is a figure which shows the example of the sentence of the affirmative sentence of FIG. 5, and the question sentence of FIG. 6 using another ending mark.

Explanation of symbols

1 Operation unit 2 Display unit 3 Sound generation unit 4 Controller 7 Speaker

Claims (9)

A sentence generation device for generating sentence data,
First storage means for storing basic sentence data including basic contents of a sentence and an ending mark indicating the ending position of the sentence;
Second storage means for storing a plurality of ending data indicating endings corresponding to different characters;
Reading means for selectively reading one ending data of the plurality of ending data from the second storage means;
Document data creating means for creating the sentence data by inserting the first ending data at the ending position indicated by the ending mark of the basic sentence data stored in the first storage means. A sentence generator. The ending mark indicates the type of sentence style indicated by the basic sentence data,
2. The sentence generating apparatus according to claim 1, wherein the reading unit selectively reads the one ending data corresponding to the type of sentence style indicated by the ending mark of the plurality of ending data. Selecting means for selecting any one of a plurality of formalities indicating different degrees of formalities;
2. The sentence generation apparatus according to claim 1, wherein the reading unit selectively reads the one ending data corresponding to the one degree of formalism among the plurality of ending data.   4. The sentence generation according to claim 3, wherein the reading means selectively reads the one ending data corresponding to the one formality and the initially set character among the plurality of ending data. apparatus.   The sentence generating apparatus according to claim 1, wherein the ending mark is arranged at a ending position of a sentence indicated by the basic sentence data.   2. The sentence generation apparatus according to claim 1, wherein the ending mark is formed by continuously arranging sentence end symbols for each type of sentence style indicated by the basic sentence data.   2. The sentence generation apparatus according to claim 1, wherein the selection unit calculates the degree of case 1 by applying an initial case form index to a predetermined random function. A sentence generation method for generating sentence data,
A first storage step of storing, in a first storage means, basic sentence data including a basic content of the sentence and an ending mark indicating the ending position of the sentence;
A second storage step of storing in the second storage means a plurality of ending data indicating endings corresponding to different personalities;
A reading step of selectively reading one ending data of the plurality of ending data from the second storage means;
A document data creation step of creating the sentence data by inserting the first ending data at the ending position indicated by the ending mark of the basic sentence data stored in the first storage means. A sentence generation method. A computer-readable program for executing a sentence generation method for generating sentence data,
A first storage step of storing, in a first storage means, basic sentence data including a basic content of the sentence and an ending mark indicating the ending position of the sentence;
A second storage step of storing in the second storage means a plurality of ending data indicating endings corresponding to different personalities;
A reading step of selectively reading one ending data of the plurality of ending data from the second storage means;
A document data creation step of creating the sentence data by inserting the first ending data at the ending position indicated by the ending mark of the basic sentence data stored in the first storage means. Program.

Images