JPS60176167A - Voice input type dictionary retrieving device - Google Patents

Abstract

PURPOSE:To obtain an appropriate KANJI (Chinese character) conversion output through reference to a conversion deciding table even when a character string including KANA (Japanese syllabary) hard to discriminate a double consonant or normal one is inputted by providing a conversion deciding table. CONSTITUTION:A conversion deciding table 4 stores characters followed by a double consonant in terms of self-dependent character instead of exoticisms. KANA strings served for sentence generation are character-encoded through an external input device and inputted after they are segmented at every word as pretreatment. The KANA string is provisionally given to a consultation control part 3 through a temporary memory device 1 and a code conversion part 2. When the inputted KANA string includes a normal consonant ''tsu'', the code conversion part 2 investigates characters followed by the normal consonant by utilizing the table 4, and generates an auxiliary KANA string in which a normal consonant ''tsu'' is replaced with a double consonant ''tsu'' only when it can be ''tsu'' of a double consonant. Thus when a character string including KANA ''tsu'' hard to discriminate a double consonant or normal one is inputted, an appropriate KANJI conversion output can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a dictionary search device for a kana-kanji conversion device. , relates to a voice input dictionary search device that enables appropriate conversion processing for input character strings that include the direct sound "tsu" even if the direct sound "tsu" is input without using an auxiliary input means. It is.

ゝConventional structure and problems thereof In recent years, text creation devices such as Japanese word processors have become popular. This type of device generally creates a Japanese sentence by inputting the desired sentence in kana characters and converting it word by word into a string of characters mixed with kanji.

This kana-kanji conversion is performed by searching for character strings that are registered in the word dictionary in advance and include kanji that respectively correspond to the kana character strings. A character string including kanji is given corresponding to the p1 kana character string.

Therefore, if a kana character string is incorrectly input, the kanji conversion will not be performed or an incorrect kanji conversion character string will be output.

However, among Japanese monosyllables, there are some syllables that are difficult to input by pronouncing them as written, and conventionally require auxiliary input means. The consonant "tsu" is a silent interval in pronunciation, and it is possible to detect it as a pause interval when the target is a completely continuous utterance, but the accuracy is low and it is difficult to detect monosyllabic voice input. In order to make it a target, it was necessary to perform an operation to indicate that it is a lowercase letter using an auxiliary input means, such as a manual keystroke.

Purpose of the Invention The purpose of the present invention was made in consideration of the above-mentioned circumstances.
Even if a string containing
To provide a highly practical voice input type dictionary search device capable of finding an appropriate kanji conversion output by checking the kana characters following the kanji character without performing unnecessary dictionary searches. .

Structure of the Invention The dictionary search device of the present invention includes a word dictionary that stores a plurality of words including kanji in correspondence with character strings representing their pronunciations;
a temporary storage device that temporarily stores the input kana character string;
When the input kana character string includes the direct sound "tsu", the character following the direct sound "tsu" is checked and only if it can be a consonant "tsu", the direct sound "tsu" is changed to a consonant. A code conversion unit that generates an auxiliary board name character string replaced with "tsu", and a code converter that searches the word dictionary for the input kana character string and the auxiliary board name character string to find words that include the kanji shown in the input kana character string. It is composed of a dictionary lookup control section to obtain the consonant "
This makes it easy to input ``So〉'' and to search in a dictionary.

DESCRIPTION OF EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic diagram showing the main parts of a voice input type dictionary search device. In FIG. 1, 1 is a temporary storage device that temporarily stores the input kana character string, and 2 is a temporary storage device that stores the input kana character string when the input kana character string includes the direct sound "tsu". Check using Conversion Judgment Table 4 to find that the consonant "tsu" is not the consonant "tsu".
A code converter that generates the auxiliary board name string replaced with
Reference numeral 3 denotes a dictionary lookup control unit that searches the word dictionary 5 using the input kana character string and the auxiliary board name character string to obtain words containing the kanji characters shown in the input kana character string.

The word dictionary 5 stores a plurality of words including kanji in correspondence with character strings representing their pronunciations. The conversion determination table 4 also stores hanging characters following consonants in independent words that are not loanwords.

The operation of the voice input type dictionary search device of this embodiment configured as described above will be explained below.

FIG. 2 is a flowchart for explaining the operation of the apparatus of the present invention.

The kana character string used for creating sentences is inputted after being converted into character codes through an input device (not shown), and subjected to preprocessing such as being separated into words. This input device is
It is composed of a keyboard device such as a kana keyboard, or a voice input recognition device (steps n2 to n4).

The input kana character string thus input is once stored in the temporary storage device 1 (step ns), and then provided to the dictionary lookup control section 3 via the code conversion section 2.

The code conversion unit 2 is equipped with a conversion determination table 4 that stores characters that can follow consonants in independent words that are not loanwords, and includes a conversion determination table 4 that stores characters that can follow consonants in independent words that are not loanwords. is included, and the character string is an independent word that is not a loan word, the character string following the direct sound "tsu" is compared with the conversion judgment table 4, and only when the conditions are satisfied, the direct sound Code conversion is performed by replacing "so" with the consonant "n".

Here, the regularity of character strings that can be consonants stored in the conversion determination table 4 is ``In independent words that are not loanwords, consonants can only appear before the consonant 'KJrTJrsJrPJ or at the end of an independent word.'' That is what it is.

Therefore, when the above conditions are not satisfied, the code converter 2
does not replace the direct sound "tsu" in the input kana string with the consonant sound "tsu".

On the other hand, if the character string is a foreign word, code conversion is performed by unconditionally replacing the direct sound "tsu" in the character string with the consonant sound "tsu" (steps n6 to n s ).

Examples of this sentence are ``May'', ``murderous intent'', ``just now'', and ``
``Question'' and ``Bed.'' The consonant “tsu” is the direct sound “tsu”
When inputting with , each input kana string is "Satsuki"
, "Satsuki", "Satsuki".

``Shitumon'' and ``bed''. When "Satsuki" is input, the consonant rKJ following "tsu" can be a consonant following the consonant according to conversion determination table 4, so the auxiliary board name character string "Satsuki" is generated. Next, when "shitsumon" is input, the consonant rMJ following "tsu" does not correspond to the consonant following a consonant, so no auxiliary board name character string is generated. However, when "bed" is input, since "bed" is a loan word, the auxiliary board name string "bed" is unconditionally input.
will generate (step n 9) 0 and
The auxiliary board name character string generated by this code conversion is also provided to the dictionary lookup control section 3 in the same manner as the input kana character string.

As a result, when it cannot be determined whether a character input with the direct sound "tsu" is a direct sound or a consonant, the regular search for character strings that can be a consonant will replace the conventional search for both character strings. If you use , you only need to search for the ``tsu'' part of Naoto, and you won't have to do any unnecessary searches.

On the other hand, the word dictionary 5 stores and registers a plurality of words in correspondence with words consisting of character strings including kanji and kana character strings indicating their pronunciations. These pieces of word information are registered and set in advance, and are registered by sequentially assigning addresses from, for example, the term "A" to the term "N". For such a word dictionary 6, the dictionary lookup control unit 3 first sequentially compares the given input kana character string with the kana character string in the dictionary to detect a match (step n,.). Then, character strings containing kanji in the word dictionary for which a match has been detected are selectively read out. At this time, if the input kana character string does not include the character "tsu", the selected character string mixed with kanji will be output as is (kanji conversion output data). In addition, input character string 91. If A does not include the character "tsu" and the corresponding word cannot be found in the word dictionary, the input kana character string is output as is. On the other hand, if the character "tsu" is included in the input kana character string, the dictionary lookup control unit 3 first searches the word dictionary for the input kana character string, and then satisfies the conditions of the conversion determination table 4. In this case, a word dictionary is searched for the auxiliary board name character string created by character code conversion. Then, conversion outputs consisting of characters mixed with kanji found by these two searches are obtained as conversion candidates. At this time, if the corresponding word cannot be found in the word dictionary even after both searches, the input kana character string is outputted (steps 11 to n13).Then, the correct one of the conversion candidate outputs is output. The selection is made, for example, by displaying this on a display and in response to an operator's instructions, based on context analysis, or by referring to information on the frequency of appearance in the text.

In this way, in this device, even if the consonant "tsu" is replaced with the direct sound "tsu" and input, this can be efficiently converted into a correct Kanji character.

Effects of the Invention As is clear from the above explanation, the present invention enables accurate input of the consonant "tsu", which is difficult to input in terms of pronunciation, even when inputting the direct sound "so" when inputting kana character strings by voice. It makes it possible to create Japanese sentences by converting kanji into kanji, and the effect is very large.

Furthermore, the device configuration is extremely convenient in practice, since it is sufficient to generate a corrected kana character string by referring to a conversion determination table. Therefore, it is possible to perform dictionary searches extremely easily and effectively.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram showing the main parts of a voice input dictionary search device according to an embodiment of the present invention, and FIG. 2 is a flowchart explaining its operation. 1... Temporary storage device, 2... Code converter, 3... Dictionary lookup control unit, 4... Conversion determination table, 5... Word dictionary. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
Figure 2

Claims (1)

[Claims] a word dictionary that stores a plurality of words including kanji in correspondence with character strings representing their pronunciations; a temporary storage device that temporarily stores input kana character strings; When the direct sound [tsu] is included, the direct sound [tsu
a code conversion unit that generates an auxiliary board name character string in which "tsu" is replaced with a consonant ")"; and a code converter that searches the word dictionary for the input kana character string and the auxiliary board name character string to find the characters indicated in the input kana character string. A voice input type % expression % characterized by comprising a dictionary lookup control unit for obtaining words including kanji.