JP2013068667A - Information processing apparatus, information processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing apparatus, an information processing method and a program in which, with respect to an entry which has been selected by a user once, if the entry appears in an example sentence of another entry, it is read aloud according to a phonetic symbol in electronic dictionary content.SOLUTION: An information processing apparatus that reads electronic dictionary content aloud, includes: temporary inhibition means that temporarily prohibits a word in the same notation as an entry from being used in the dictionary; registration means that registers an entry, a phonetic symbol, and a speech synthesis phonetic symbol to the dictionary; and inhibition release means that allows a word unusable made temporarily by the temporary inhibition means to be used in the dictionary.

Description

  The present invention relates to an information processing apparatus, an information processing method, and a program.

  In the electronic dictionary, there is a technique described in Patent Document 1 as a technique for synthesizing speech according to phonetic symbols described in the electronic dictionary content when reading an entry word or an example sentence including the entry word in the electronic dictionary content. In the technique of Patent Literature 1, when a headword and headword explanation information are displayed, phonetic symbols and part-of-speech information are extracted from the headword explanation information. When an example sentence to be read is designated by the user, the extracted phonetic symbol and part-of-speech information of the extracted headword are inserted into the example sentence and passed to the speech synthesis means. The voice synthesizing means synthesizes the headword in the example sentence specified by using the phonetic symbol and part-of-speech information of the inserted headword by reading according to the phonetic symbol of the electronic dictionary content.

Japanese Patent No. 4767869

However, in Patent Document 1, there is a problem that, regarding the headword once selected by the user, when the headword appears in an example sentence of another headword, it cannot be read out as the phonetic symbol of the electronic dictionary content.
The present invention has been made in view of such a problem. When the headword appears in an example sentence of another headword with respect to the headword once selected by the user, it is read out as the phonetic symbol of the electronic dictionary content. For the purpose.

  Therefore, the present invention provides an information processing apparatus that reads out electronic dictionary content, a search unit that searches for electronic dictionary content based on an input headword, and a headword that is a search result of the search unit. And display means for displaying the explanation information of the headword on a display device, phonetic symbol specifying means for specifying the phonetic symbol of the headword that is a part of the headword explanation information, and the phonetic symbol specifying means Conversion means for converting the specified phonetic symbol into a phonetic synthesis phonetic symbol; temporary suppression means for temporarily disabling a word having the same notation as the headword in the dictionary; the headword; and the phonetic symbol; , Temporarily registered by the registration means for registering the phonetic symbol for speech synthesis in the dictionary, the speech synthesis means for synthesizing the reading portion, and the temporary suppression means. The serial word having a derepression means available in the dictionary.

  According to the present invention, when an entry word appears in an example sentence of another entry word, the entry word once selected by the user can be read out according to the pronunciation symbol of the electronic dictionary content.

FIG. 3 is a diagram illustrating an example of a software configuration of the information processing apparatus (electronic dictionary apparatus) according to the first embodiment. It is a figure which shows an example of the hardware constitutions of information processing apparatus. It is a figure which shows an example of the headword of the electronic dictionary content. It is a flowchart which shows the flow of basic information processing. It is a figure which shows an example of the software configuration etc. of the information processing apparatus (electronic dictionary apparatus) of Embodiment 2. It is a flowchart which shows the flow of the information processing when the upper limit of the number of words which can be registered into a user dictionary is considered. It is a flowchart which shows the flow of the process of difference cost calculation in step S701. It is a figure which shows an example of the software configuration etc. of the information processing apparatus (electronic dictionary apparatus) of Embodiment 3. It is a flowchart which shows the flow of a process in the case of registering all the words of the same description with a headword in a user dictionary.

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

<Embodiment 1>
FIG. 1 is a diagram illustrating an example of a software configuration of the information processing apparatus (electronic dictionary apparatus) according to the first embodiment.
The headword designation unit 101 designates a headword.
The headword search unit 102 searches the electronic dictionary content 112 for the headword specified by the headword specifying unit 101, and obtains the description information corresponding to the headword as a search result.
The display unit 103 displays the search result of the headword search unit 102.
The reading location selection unit 104 selects a reading location from the explanation information.
The part-of-speech extraction unit 105 extracts the part-of-speech corresponding to the headword included in the reading location from the explanation information.
The phonetic symbol extraction unit 106 extracts phonetic symbols corresponding to headwords included in the reading portion from the explanation information.
The phonetic symbol conversion unit 107 converts the extracted phonetic symbols into speech synthesis readings.

When a word having the same notation as the headword is registered in the user dictionary 109, the same notation word temporary suppressing unit 108 temporarily deletes all the same notation words from the user dictionary 109. The same notation word temporary suppression unit 108 holds the word deleted from the user dictionary so that it can be registered again.
The user dictionary 109 is a speech synthesis dictionary that can dynamically register and delete words. When performing language analysis for speech synthesis, a word registered in the user dictionary is selected in preference to the synthesis language dictionary 113.
The user dictionary search unit 110 searches the user dictionary 109 for words, and obtains information necessary for speech synthesis such as word notation, part of speech, and reading as a search result.
The user dictionary deletion unit 111 deletes registered words from the user dictionary 109.
The electronic dictionary content 112 is an electronic content such as an English-Japanese dictionary, a Japanese-English dictionary, a Chunichi dictionary, or a Japanese-German dictionary installed in the electronic dictionary.
The synthesis language dictionary 113 is a static dictionary prepared in advance and holds information necessary for speech synthesis, such as word notation, parts of speech, and reading.

The synthesizing language dictionary search unit 114 searches the synthesizing language dictionary 113 for a word, and obtains information necessary for speech synthesis such as word notation, part of speech, and reading as a search result.
The unknown word reading unit 115 is used for words that are not registered in the synthesis language dictionary 113 in the user dictionary 109, and estimates readings using the notation as input.
The headword registration unit 116 sets the headword specified by the headword specification unit 101, the part of speech extracted by the part of speech extraction unit 105, and the reading converted by the phonetic symbol conversion unit 107 as a set, through the user dictionary registration unit 117. 109 is registered.
The user dictionary registration unit 117 registers words in the user dictionary 109.
The speech synthesizer 118 reads the reading portion selected by the reading portion selection unit 104 using the synthesis language dictionary 113, the user dictionary 109, and the unknown word reading unit 115, and generates synthesized speech according to the reading. And output.
The same notation word suppression release unit 119 registers the same notation word temporarily deleted by the same notation word temporary suppression unit 108 in the user dictionary 109 again.

FIG. 2 is a diagram illustrating an example of a hardware configuration of the information processing apparatus.
The CPU 202 generally controls each device connected via the system bus 205. The CPU 202 reads and executes processing steps and programs stored in the read-only memory (ROM) 203 or the flash memory (FM) 211. A random access memory (RAM) 204 is used as a temporary storage area such as a main memory and work area of the CPU 202 that can be accessed at high speed.
Here, the FM 211 stores, as programs, processes that can be executed by a computer in an operating system, applications, and the like, and stores device drivers and the like that are executed by the CPU 202.
ROM 203 stores invariant processing steps that can be executed by the computer for basic system functions such as basic I / O, startup, or keystroke acceptance from keyboard 208.
A keyboard controller (KBC) 206 controls an instruction input from a pointing device 207 such as a keyboard (KB) 208, a mouse, a touch panel, and a laser pointer.
A display controller (DSPC) 209 controls display on the display (DSP) 210.
The electronic dictionary content 112, the synthesis language dictionary 113, and the user dictionary 109 are configured in an FM 211 or the like.
When the CPU 202 executes processing based on a program, the software configuration of the information device, processing of each step of the flowchart described later, and the like are realized.

FIG. 3 is a diagram illustrating an example of a headword of the electronic dictionary content 112.
Reference numeral 501 denotes a headword, which in this example is the word today.
Reference numeral 502 denotes headword description information. The explanation information includes phonetic symbols 503, parts of speech 504 and 505, and example sentences.
Reference numeral 506 denotes a reading portion designated by the user. This example shows an example in which the reading-out part is highlighted.

With reference to the flowchart of FIG. 4, basic processing contents in which the present embodiment having the configuration shown in FIGS. 1 and 2 performs reading, registration of a headword to the user dictionary, and deletion will be described.
In step S301, the headword designating unit 101 receives a headword input by the user, and proceeds to step S302.
In step S <b> 302, the headword search unit 102 searches for headwords held by the headword specification unit 101 for the electronic dictionary content 112. The display unit 103 displays the headword and the headword explanation information as a search result of the headword search unit 102 on the display 210, and proceeds to step S303.
In step S303, the reading part selection unit 104 accepts the reading part input by the user, and the process proceeds to step S304. It is assumed that the reading portion is a part of the description information of the headword.
In step S304, if the user presses the speech synthesis button, the process proceeds to step S305, and if not, the process proceeds to step S304.

A headword is registered in the user dictionary 109 in steps S305 to S309.
In step S305, the part-of-speech extraction unit 105 identifies the part-of-speech of the headword (part-of-speech identification) for the headword in the reading-out part held by the reading-out part selection unit 104. More specifically, the part-of-speech extraction unit 105 extracts a part related to the part-of-speech from the description information of the headword and identifies the part-of-speech. When there are a plurality of parts of speech for one headword, the part of speech can be specified on the basis of the position of the reading portion in the explanation information of the headword. When the part of speech is specified, the process proceeds to step S306.
In step S306, the phonetic symbol extraction unit 106 specifies the phonetic symbol of the headword (phonetic symbol specification) for the headword in the reading portion held by the reading portion selection unit 104. More specifically, the phonetic symbol extraction unit 106 extracts a portion related to the phonetic symbol from the description information of the headword and identifies the phonetic symbol. If there is a plurality of phonetic symbols for one headword, the phonetic symbol extraction unit 106 specifies the corresponding phonetic symbol based on the part of speech specified in step S305. If a phonetic symbol is specified, it will move to step S307.

In step S307, the phonetic symbol converter 107 converts the phonetic symbols into phonetic symbols for speech synthesis. Since the phonetic symbols extracted by the phonetic symbol extraction unit 106 are based on the phonetic symbols and phonetic symbol system defined in the electronic dictionary content, it is necessary to convert them into phonetic symbols for speech synthesis. For this reason, a phonetic symbol conversion table is prepared in advance. If the phonetic symbol is converted, the process proceeds to step S308.
In step S <b> 308, the same notation word temporary suppression unit 108 temporarily disables the word having the same notation as the headword in the user dictionary 109. Here, in order to realize temporary disabling, the same notation word temporary suppression unit 108 deletes a word from the user dictionary. If words with the same notation but different phonetic symbols have already been registered in the user dictionary, simply registering the headword in the user dictionary will result in a linguistic analysis of the reading-out part, and the desired reading (that is, the headword (Reading) may not be selected. Here, a word having the same notation as the headword is temporarily deleted from the user dictionary so that reading of the headword is surely selected. More specifically, the user dictionary search unit 110 searches the user dictionary 109 for a headword. The user dictionary deletion unit 111 deletes words that have been found in the search from the user dictionary. The user dictionary deletion unit 111 deletes all the words that have been hit in the search. Since the deleted word is re-registered in the user dictionary in step S312 after the speech synthesis in step S310, the same notation word temporary suppression unit 108 holds it. The word having the same notation is temporarily disabled in the user dictionary, and the process proceeds to step S309.

In step S309, the headword registration unit 116 registers headwords in the user dictionary 109. More specifically, the user dictionary registration unit 117 registers the notation of the headwords held by the headword search unit 102, the part of speech held by the part of speech extraction unit, and the phonetic symbols held by the phonetic symbol conversion unit 107 in the user dictionary 109. . After registration, the process proceeds to step S310.
In step S310, the speech synthesizing unit 118 synthesizes the speech that is read out by the speech selection unit 104. When the synthesized speech is output, the process proceeds to step S311.
In step S311, if the same notation word is temporarily suppressed in step S308, the process proceeds to step S312, and if not temporarily suppressed, the process ends.
In step S312, the same written word suppression canceling unit 119 registers the same written word in the user dictionary 109 again to release the suppression of the same written word temporarily suppressed in step S308. More specifically, the user dictionary registration unit 117 registers the temporarily deleted words held by the same notation word temporary suppression unit 108 in the user dictionary 109. When there are a plurality of temporarily deleted words, the same written word suppression cancellation unit 119 registers all of them. After registration, the process ends.

The effect of using a user dictionary for speech reading in this embodiment will be described using a more specific example. It is assumed that the word blog (pronunciation b 1 a: g) is registered in the electronic dictionary content 112, while not registered in the synthesis language dictionary 113. Further, when reading processing by the unknown word reading unit 115 is performed, the word blog is given a reading bla: dzI.
The user searches, displays, and synthesizes a word blog using an electronic dictionary, and then searches, displays, and synthesizes another word, for example, delete. It is assumed that the word blog is included in the reading-out part when speech synthesis is performed with the word delete.
When the user searches for the word blog in the electronic dictionary, displays it, and synthesizes the speech (steps S301 to S310), the word blog is registered in the user dictionary 109 along with the phonetic symbols and parts of speech described in the description information of the blog headword. Is done. Therefore, at the time of speech synthesis, blog registered in the user dictionary is selected, and speech synthesis is performed with pronunciation (b 1 a: g) according to the electronic dictionary content. For example, the read-out part is I read his blog everday. Then, the word blog is correctly pronounced as b l a: g.
The same notation word is re-registered according to the presence or absence of the temporarily deleted word (steps S312 and S313), and the process is terminated. As a result, the word blog remains registered in the user dictionary.

Next, the user searches for the word delete in the electronic dictionary, displays it, and performs speech synthesis. I read my blog. And Since the headword is delete, the word delete is registered in the user dictionary 109. When speech synthesis is performed, not only the word delete but also the word blog are correctly synthesized with pronunciation according to the electronic dictionary content. This is because the word blog was registered in the user dictionary when the user synthesized the word blog last time. The headwords that have been speech-synthesized in this way are automatically registered and stored in the user dictionary, so that even if they appear in an example sentence of another headword, it is possible to synthesize voice according to the phonetic symbol of the electronic dictionary content. become.
In the present embodiment, the case where the same notation word temporary suppression unit 108 deletes the same notation word from the user dictionary in order to temporarily disable the word in the user dictionary has been described. However, the present invention is not limited to this. . For example, as a function of the user dictionary, it may be possible to control the usable / unusable state for each registered word. In this case, the same notation word temporary suppression unit 108 makes the same notation word registered in the user dictionary unusable via a user dictionary state setting unit (not shown) that sets the word state of the user dictionary. Set. It is assumed that the words set to be unusable are not used in the language analysis of speech synthesis, and the words set to be usable are used in the language analysis of speech synthesis. Similarly, the same notation word suppression release unit 119 sets the same notation word temporarily disabled by the same notation word temporary suppression unit 108 to be usable in the user dictionary. It should be noted that not only the binary state, such as usable or unusable, may be configured to set the priority. When it is desired to suppress the use in the user dictionary, the priority may be sufficiently lowered, and conversely, when it is desired to cancel the suppression, the priority may be returned to a predetermined value.

<Embodiment 2>
Generally, there is an upper limit on the number of words that can be registered in the user dictionary. That is, after registering a word up to the upper limit, in order to read out the headword that is the premise of the present embodiment as pronounced in the electronic dictionary content, it is necessary to delete the registered word in order to register a new word. is there. For that purpose, it is necessary to decide which word is deleted. As a simple method, the oldest registration date or the least frequently used order is effective.
Hereinafter, as another effective method, a deletion word determination method focusing on a difference between a phonetic symbol of electronic dictionary content and a phonetic symbol of speech synthesis will be described.
FIG. 5 is a diagram illustrating an example of a software configuration of the information processing apparatus (electronic dictionary apparatus) according to the second embodiment.
In FIG. 5, the same modules as those in FIG.
The difference cost calculation unit 601 calculates the difference between the speech synthesis reading converted by the phonetic symbol conversion unit 107 and the reading registered in the synthesis language dictionary 113 or the reading estimated by the unknown word reading unit 115.
The deletion word determination unit 602 determines a word to be deleted according to the difference cost calculated by the difference cost calculation unit 601 when there is not enough space in the user dictionary.

FIG. 6 is a flowchart showing the flow of processing when the upper limit of the number of words that can be registered in the user dictionary is considered. In FIG. 6, steps S301 to S307, S309, and S310 are the same as those in FIG.
In step S308, the same notation word temporary suppression unit 108 temporarily deletes the same notation word from the user dictionary, and the process proceeds to step S701.
In step S701, the difference cost calculation unit 601 calculates the difference cost of the headword. The difference cost is used to determine a word to be deleted when the number of words registered in the user dictionary reaches the upper limit in steps S703 and S906. The difference cost is obtained by quantifying a difference between reading of a headword in electronic dictionary content and reading of a headword by speech synthesis. Here, it is defined by the difference between the phonetic symbol of the headword of the electronic dictionary content and the phonetic symbol output in the middle of speech synthesis. There are two possibilities for phonetic symbols output by speech synthesis in the middle. The case where the headword is registered in the synthesis language dictionary 113 and the case where it is not. If it is registered in the synthesis language dictionary, the phonetic symbols of the synthesis language dictionary are used. On the other hand, if it is not registered in the synthesis language dictionary, the unknown word reading unit 115 uses a phonetic symbol estimated from the entry word notation.

Hereinafter, the flow of the difference cost calculation process will be described with reference to FIG.
FIG. 7 is a flowchart showing the flow of the difference cost calculation process in step S701.
In step S401, the synthesis language dictionary search unit 114 searches for a headword in the synthesis language dictionary 113, and proceeds to step S402.
In step S402, the synthesis language dictionary search unit 114 proceeds to step S404 if the search is hit, that is, if the headword is registered in the synthesis language dictionary 113. If it is not registered, the synthesis language dictionary search unit 114 proceeds to step S403.
In step S403, the unknown word reading unit 115 estimates phonetic symbols using the notation of the headword as input. The phonetic symbol is estimated and the process proceeds to step S404.
In step S404, the difference cost calculation unit 601 calculates the difference cost between the phonetic symbol of the headword of the electronic dictionary content (the phonetic symbol for speech synthesis held by the phonetic symbol conversion unit 107) and the phonetic symbol obtained in steps S401 to S403. Is calculated. When the difference cost is calculated, the process ends.

Here, an example of the method for calculating the difference cost will be described. Based on the phonetic symbol of the headword of the electronic dictionary content and the phonetic symbol obtained in steps S401 to S403, it can be defined as follows.
Difference cost = (number of insertion errors + number of omission errors + number of substitution errors) / number of phonetic symbols of entry words of dictionary content A difference cost calculation example will be described using a more specific example.
Phonetic symbol of electronic dictionary contents headword b l a: g
Phonetic symbols output by speech synthesis in the middle b l a: dz I
When DP matching is performed between two phonetic symbols and the correspondence of each phonetic symbol is determined, it is as follows.
b l a: g
b l a: dz I
b, l, a: match, g and dz are substitution errors, and I is an insertion error. Therefore, the difference cost is
Difference cost = (1 + 0 + 1) /4=0.5
It becomes.
Returning to the description of FIG. 6, when the difference cost is calculated in step S <b> 701, the process proceeds to step S <b> 309.

In step S309, the headword registration unit 116 registers the headword notation, part of speech, phonetic symbol, and difference cost in the user dictionary 109. The user dictionary can hold a difference cost for each word. After registration, the process proceeds to step S310.
In step S311, if the same notation word is temporarily deleted in step S308, the process proceeds to step S312, and if not temporarily deleted, the process proceeds to step S702.
In step S312, the same notation word suppression release unit 119 registers the same notation word temporarily deleted in step S308 in the user dictionary 109 again, and the process proceeds to step S702.
In step S702, when there is no space for registering a new word in the user dictionary 109, the process proceeds to step S703, and when there is a space, the process ends.
In step S703, the deleted word determination unit 602 determines a word to be deleted from the user dictionary 109. The word to be deleted is the word with the smallest difference cost of words registered in the user dictionary 109.

Here, the fact that the difference cost is small means that the difference between reading of words in the electronic dictionary content and reading of the synthesized speech is small when speech synthesis is performed. That is, the value of registering the user dictionary is small. Conversely, a large difference cost means that when speech synthesis is performed, if the word is not registered in the user dictionary, the speech synthesis reading difference is large. Therefore, by deleting from a word having a small difference, a word having a large difference from the word reading of the electronic dictionary content is left in the user dictionary as much as possible. As a result, words with a large difference cost remaining in the user dictionary can be pronounced by speech synthesis as the electronic dictionary content is read, so speech synthesis that is as close to reading of the electronic dictionary content as possible within the limited number of registered words in the user dictionary. Can be expected to be realized.
When the word to be deleted is determined, the user dictionary deleting unit 111 deletes the word from the user dictionary 109, and the process ends.

In the present embodiment, in order to be able to register a headword in the user dictionary 109 without confirming the availability of the user dictionary 109 in step S309, a process of securing at least one vacancy is performed in steps S702 and S703. ing. Note that the availability of the user dictionary 109 is confirmed before step S309, and if there is no available space, the word with the smallest difference cost may be deleted from the user dictionary.
So far, the case where only the headword specified by the headword specifying unit 101 is registered in the user dictionary has been described. However, the present invention is not limited to this, and a word having the same notation as the headword is also registered. It shall be good.

<Embodiment 3>
Here, consider a case where there are words with the same notation and different readings. As described above, the word registered in the user dictionary 109 is selected in preference to the word registered in the synthesis language dictionary 113. If only one of the reading words is registered in the user dictionary, the reading is always adopted. Therefore, when the word appears in an example sentence of another headword, the possibility of a reading error increases. Therefore, all the same notation words are registered in the user dictionary. As a result, when language analysis is performed on the read-out part, words having a high possibility of connection before and after are selected based on the part of speech of the same notation word registered in the user dictionary, and there is a high possibility that speech synthesis is performed with correct reading. . On the other hand, when deleting, the same notation word is deleted from the user dictionary, so that words that have a high possibility of connection before and after based on the part-of-speech of the word registered in the synthesis language dictionary when language analysis is performed on the reading portion Is selected. This increases the possibility of voice synthesis with correct reading.
Hereinafter, with reference to FIG. 8 and FIG. 9, a case where a word having the same notation as the headword is also registered will be described.

FIG. 8 is a diagram illustrating an example of a software configuration of the information processing apparatus (electronic dictionary apparatus) according to the third embodiment.
In FIG. 8, the same modules as those in FIG.
The headword search unit 801 searches the description information obtained by the headword search unit 102 for words having the same notation as the headword.
The headword non-search part 802 searches the electronic dictionary content 112 for a word having the same notation as the headword.

FIG. 9 is a flowchart showing the flow of processing when registering all the headwords and the same notation words in the user dictionary.
In FIG. 9, steps S301 to S310, S312 and S70 to S703 are given the same numbers to steps performing the same processing as in FIG.
In step S311, if the same notation word is temporarily deleted in step S308, the process proceeds to step S312, and if not temporarily deleted, the process proceeds to step S901.
In step S901, the headword search unit 801 searches the headword explanation information for words having the same notation as the headword but different parts of speech and pronunciation. The headword search unit 801 holds the search result and proceeds to step S902.
In step S <b> 902, the headword non-word search unit 802 searches the electronic dictionary content 112 for words having the same notation as the headword but different parts of speech and pronunciation. Words with the same notation are often included in the description information of the same headword, but may be described as a separate headline depending on a dictionary or depending on the word. Step S902 is a process for that purpose. The headword search unit 802 holds the search result and proceeds to step S903.

In step S903, as a result of step S901 and step S902, if there are words with the same notation but different parts of speech or phonetic symbols, the process proceeds to step S904. If not, the process proceeds to step S702.
In step S904, the difference cost calculation unit 601 calculates a difference cost for each word hit in the search. For example, when three words are hit, the difference cost calculation unit 601 first obtains a difference cost for each of the three words. Further, the difference cost calculation unit 601 obtains an average value of the difference costs of the three words, and sets this value as the difference cost common to the three words. This is a device for deleting three words at the same time when deleting words having a small difference cost from the user dictionary. As described above, the word registered in the user dictionary 109 is selected in preference to the word registered in the synthesis language dictionary 113. Therefore, if the same notation word partially remains in the user dictionary, the word is always adopted, so that the possibility of reading errors increases. When deleting from the user dictionary, by deleting all the same notation words, when performing a linguistic analysis of the reading part, there is a possibility of connection before and after based on the part of speech of the word registered in the language dictionary for synthesis. A high word is selected. This increases the possibility of voice synthesis with correct reading. After calculating the difference cost, the difference cost calculation unit 601 proceeds to step S905.

In step S905, if there is not enough room for registering a new word in the user dictionary 109, the process proceeds to step S906, and if there is sufficient room, the process proceeds to step S909.
Here, if the number of identical notation words that are user dictionary registration candidates is Nc, sufficient space is Nc + 1 words. The meaning of +1 is to always ensure an empty space for one word in the user dictionary 109 in step S309 as described above.
In step S906, in order to determine the word to be deleted by the deleted word determination unit 602, the difference cost of the same notation word that is a user dictionary registration candidate is compared with the difference cost of the word registered in the user dictionary. When the empty space of the user dictionary is Nv, the number Nd of words to be deleted from the user dictionary can be expressed as follows.
Nd = Nc + 1−Nv
For comparison, the words registered in the user dictionary may be arranged in ascending order of the difference cost and compared with the Nd-th value. In order to speed up the processing, it may be sorted in advance by the value of the difference cost. After the comparison, the process proceeds to step S907.

In step S907, if the comparison shows that the difference cost of the same notation word that is a user dictionary registration candidate is smaller than the Nd-th value, the process ends. That is, since the difference cost of the same notation word which is a user dictionary registration candidate is small, it is not registered. If not, the process moves to step S908.
In step S908, the deleted word determination unit 602 determines a word to be deleted. Based on Nd obtained in step S906, at least Nd words are determined as words to be deleted from the user dictionary in ascending order of difference cost. Here, “at least” is described because the same notation word as the Nd-th word is also a deletion word. As described in step S904, the difference cost of the Nd-th word and the same notation word is the same. The user dictionary deletion unit 111 deletes the deleted word from the user dictionary 109, and the process proceeds to step S909.
In step S909, the user dictionary registration unit 117 registers the same notation word as a user dictionary registration candidate in the user dictionary 109, and the process ends.

  As described above, words having the same notation as the headword are also registered in the user dictionary, and when deleting, all the words having the same notation are deleted. As a result, even when a headword once synthesized by speech and the same notation word appear in an example sentence of another headword, only the headword is registered in the user dictionary so that it may be read as the phonetic symbol of the electronic dictionary content 112. It becomes higher than when you do.

<Other embodiments>
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, etc.) of the system or apparatus reads the program. It is a process to be executed.

As described above, according to each of the above-described embodiments, when the headword appears in an example sentence of another headword with respect to the headword selected once by the user, it can be read out as the phonetic symbol of the electronic dictionary content.
In the above-described embodiment, a case has been described in which a user dictionary is used as a dictionary for dynamically adding and deleting words and setting the state of words. In general, a user dictionary is used by a user to synthesize a desired word by speech as desired. In the above-described embodiment, the case where the electronic dictionary system uses this user dictionary has been described. However, the dictionary is not limited to a user dictionary, and any dictionary that can dynamically add and delete words and set the state of words and that selects registered words preferentially over the synthesis language dictionary. I shall do it.
Moreover, although embodiment mentioned above demonstrated the case where a user selected the reading location in detailed description of a headword, it is not limited to this. For example, when the detailed description of the headword is displayed, all the detailed description may be automatically determined as a reading portion. In addition, any method may be used, such as when the entire page is automatically determined as a reading position at the timing when the page to be displayed is switched.
In the above-described embodiment, the case where the user designates a headword or a reading position via the KBC 206 has been described. However, the present invention is not limited to this. It may be realized by any input means such as line-of-sight input or gesture input.

The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to such specific embodiments, and various modifications can be made within the scope of the gist of the present invention described in the claims.・ Change is possible.
For example, an information processing apparatus that reads out electronic dictionary content is also realized by a mobile terminal device or a computer that can display and operate the electronic dictionary content on a screen.

202 CPU, 203 ROM, 204 RAM, 211 FM

Claims (8)

An information processing apparatus that reads out electronic dictionary content,
Search means for searching electronic dictionary content based on the entered headword;
Display means for displaying a headword as a search result in the search means and description information of the headword on a display device;
Phonetic symbol specifying means for specifying the phonetic symbol of the headword of the reading portion that is a part of the description information of the headword;
Conversion means for converting the phonetic symbols specified by the phonetic symbol specifying means into phonetic synthesis phonetic symbols;
Temporary suppression means for temporarily disabling words in the same notation as the headword in the dictionary;
Registration means for registering the headword, the phonetic symbol, and the phonetic symbol for speech synthesis in the dictionary;
Speech synthesis means for synthesizing the reading location;
Suppression release means for enabling use of the word temporarily disabled by the temporary suppression means in the dictionary;
An information processing apparatus.   The information processing apparatus according to claim 1, wherein the temporary suppression unit is temporarily disabled by temporarily deleting a word having the same notation as the headword from the dictionary.   The information processing apparatus according to claim 2, wherein the suppression release unit enables the word to be used in the dictionary by registering the same notation word temporarily deleted by the temporary suppression unit in the dictionary again. Calculating means for calculating a difference cost indicating a difference between reading of a headword in the electronic dictionary content and reading of a headword by voice synthesis in the voice synthesizer;
When there is no space for registering a new word in the dictionary, a deletion unit that deletes a word having a minimum difference cost calculated by the calculation unit from the dictionary;
The information processing apparatus according to claim 1, further comprising:   The information processing apparatus according to claim 4, wherein the deletion unit also deletes from the dictionary the same notation word as the word having the smallest difference cost calculated by the calculation unit.   The information processing apparatus according to claim 1, wherein the registration unit further registers a word having the same notation as the headword together with the headword in the dictionary. An information processing method executed by an information processing apparatus that reads out electronic dictionary content,
A search step for searching electronic dictionary content based on the entered headword;
A display step of displaying on the display device the headword and the headword explanation information as a search result in the search step;
A phonetic symbol specifying step for specifying a phonetic symbol of the headword of the reading portion that is a part of the explanation information of the headword;
A conversion step of converting the phonetic symbol specified in the phonetic symbol specifying step into a phonetic symbol for speech synthesis;
A temporary suppression step of temporarily disabling words in the same notation as the headword in the dictionary;
A registration step of registering the headword, the phonetic symbol, and the phonetic symbol for speech synthesis in the dictionary;
A speech synthesis step of performing speech synthesis of the reading location;
A suppression release step of making the word temporarily disabled in the temporary suppression step usable in the dictionary;
An information processing method including: To a computer that reads out electronic dictionary content,
A search step for searching electronic dictionary content based on the entered headword;
A display step of displaying on the display device the headword and the headword explanation information as a search result in the search step;
A phonetic symbol specifying step for specifying a phonetic symbol of the headword of the reading portion that is a part of the explanation information of the headword;
A conversion step of converting the phonetic symbol specified in the phonetic symbol specifying step into a phonetic symbol for speech synthesis;
A temporary suppression step of temporarily disabling words in the same notation as the headword in the dictionary;
A registration step of registering the headword, the phonetic symbol, and the phonetic symbol for speech synthesis in the dictionary;
A speech synthesis step of performing speech synthesis of the reading location;
A suppression release step of making the word temporarily disabled in the temporary suppression step usable in the dictionary;
A program for running

Images