JPS59127150A - Sentence reading and checking device - Google Patents

Abstract

PURPOSE:To improve the efficiency of collation by constituting a character train unit dictionary for each type of character train and using selectively these dictionaries in response to the types of sentences. CONSTITUTION:The character trains which are possibly used in the sentences are stored together with the advance of reading into character train unit dictionaries 26-1-26-n respectively. In this case, both the dictionary and the character train to be stored in said dictionary are decided by the types (words, paragraphs, etc.) of character trains. A sentence read out of a sentence file 27 through a sentence reading part 28 is converted by a reading converting part 30 every character train unit of words, etc. and in accordance with the reading way given from one of the dictionaries 26-1-26-n which is selected by a dictionary selecting part 25 and then undergoes the synthesization of voice through a voice synthesizing part 31. The synthesized sentence is outputted in sound through a voice output part 32, and the sentence is read in voice.

Description

[Detailed Description of the Invention] Technical Field of the Invention The present invention uses language processing used in printing, publishing, and other operations to automatically detect and correct errors such as typos and omissions in sentences. The present invention relates to a text reading inspection device that can be applied to a text abnormality detection and correction device, etc.

(B) Technical background Many parts of printing, publishing, and other operations have been mechanized and automated. However, since humans are the ones who think about sentences, it is inevitable that errors will occur during the process of inputting the sentences into a machine. This may be a typographical error caused by the writer's misunderstanding, or a typo or omission during input. Furthermore, when publishing newspapers, etc., the range of terms used is limited in accordance with social common sense, and at present it is essential for humans to inspect and proofread text input by machines. This manual inspection and calibration requires a great deal of effort and time. In addition, manual proofreading is also prone to errors and omissions, and double or triple inspection and proofreading may be required in some cases.

Automatic inspection of texts to perform such inspection and proofreading,
Correcting devices have also been proposed, and in the past, in English and other languages, there was a device that detects words that are not in the dictionary (with a high possibility of being misspelled) by matching the spelling of words in a sentence with words in the dictionary. However, the above method cannot be used in languages such as Japanese, where word boundaries are unclear. For this reason, the present inventors have proposed an abnormality inspection and correction device for sentences suitable for language processing such as Japanese.

(Q. Prior Art and Problems) The above-mentioned text inspection and correction device (in this case, it not only displays the character string or text being inspected, how to correct the result of checking with a dictionary, etc. on the display, but also displays it in audio). It is desirable to have a function that converts and reads out aloud.
It is also used to compare a character code string entered into a word processor, etc.) with a manuscript.

In such a text reading test device, the words that can be used may be significantly limited depending on the type of text. In such cases, it is uneconomical to memorize all the words in the dictionary.
Therefore, it is desirable to prepare an efficient dictionary.

(To) Purpose of the Invention Accordingly, the purpose of the present invention is to provide a text reading/reading inspection device with an efficient dictionary structure, which can divide a dictionary according to word type and use them selectively depending on the text. It is about providing.

A text reading inspection device that has a character string unit dictionary storing information and reads the text aloud while comparing it with a character string in a sentence, has a plurality of said character string unit dictionaries, and has a plurality of character string units. The plurality of character string unit dictionaries are selectively selected by the selection means depending on the type of the text. The present invention is characterized in that the selection means selects and uses the plurality of character string unit dictionaries with priority.

■ Embodiment of the Invention Fig. 1 shows an example of a text inspection/correction device proposed by the inventors to which the text reading device of the present invention is applied, and Fig. 2 shows an example of the text reading device of the present invention. FIG.

In Figure 1, 1 is an audio output device such as a speaker or headphone, 2 is a character display device such as a Japanese display, and 3 is a character display device such as a Japanese display.
Character input device such as a kanji input keyboard, 4 is a speech synthesis device, 5 is a character display control unit, 6 is a system control unit that controls the integration of this system, 7 is a modified character string input unit, 8 is an arbitrary word voice parameter 9 is a synthesis unit; 9 is an arbitrary word speech parameter dictionary; 10 is an automatically corrected character string determining unit; 11 is a character string correction unit; 12 is a text reading unit; 13 is a text storage unit; 14
15 is a text analysis result storage unit, 16 is an abnormal character string detection unit, 17 is a dictionary extraction unit, 18 is a character string,
It is a reading dictionary.

First, the text to be inspected is read out from the text storage device 13 by the text reading section 12 and sent to the abnormal character string detection section 16 and the character display control section 51. Sentence P storage device I
Reference numeral 3 denotes a magnetic disk or a semiconductor storage device, which stores sentences (character code strings) as well as sentence numbers, page numbers, line numbers, etc. necessary for managing the sentences.

The text reading unit 12 sequentially reads out the text to be inspected from the text tracing device 13 in units of sentences or small units delimited by punctuation marks, commas, parentheses, etc. send to

The sentence input to the abnormal character string detection section is compared with the character string read out from the character string/reading dictionary 18 by the dictionary extraction section 17.

The character string/reading dictionary 18 contains character strings (words, compound words) that are likely to appear in a sentence and their connection relation information (permitted in a sentence/
combinations and permissible combinations), character strings that are unlikely to appear in sentences (wrong words), how to correct them (correct word strings), and audio information for use in reading numbers ( I remember one or more groups of words (reading, word accent type).

Equipped with an address table, etc. for quickly reading out necessary information from the dictionary 17 character string/reading dictionary 18, it is equipped with an address table etc. for quickly reading out the necessary information from the dictionary 18, and matches the character string in the text according to instructions from the abnormal character string detection unit 16 or reading analysis unit 14. The information regarding the character string ζ in the dictionary is read from the dictionary 18 and sent to the abnormal character string detection section 16 or the reading analysis section 14 or the automatically corrected character string determination section 10.

The abnormal character detection section 1°6 reads out the character string of the test target sentence read from the text storage device 13 and the dictionary entry section 17 (
From this, it is determined whether the combination of character strings read out matches, whether the combination is allowed or not, and whether a character string that matches the incorrect word exists in the test target sentence. , and sends the results to the automatically corrected character string determination section 10 and character display control section 5.

To check whether the above combinations match, first extract all character strings in the dictionary that match partial character strings in the sentence, and check whether the combinations match the sentence in just the right amount or not. A method for determining whether or not a silk matching exists can be applied, and any of these methods can be easily realized by a trader.

Also, whether the above combination is allowed or not is checked for each character string in the dictionary based on its forward connection property, backward inscription #4?
The combination of the backward connection characteristic and the forward connection characteristic that can be twisted is stored separately, and at the boundary of each character string, the backward connection characteristic of the preceding character string, A method of determining whether the forward connection characteristic of the trailing character string is allowed can be applied.

In addition, the abnormal character string detection unit 16 sends the sentence analysis results (information such as how to combine dictionary character strings that match the sentence) to the sentence analysis result storage unit 51 for use in the reading analysis unit 14, if necessary. Shi, I remember it.

As a result of the above judgment, the abnormal character string detection unit 16 detects the entire sentence whose combination does not match or the partial character string in the sentence which cannot be expressed by the combination f as abnormality of the sentence. , character display control section 5
Send to.

The automatic correction character string determination unit uses the character display control unit 5 as necessary based on the incorrect word input from the abnormal character string detection unit 16 and the correction method input from the dictionary search unit 17.
Sends the correction method, displays it on the character display device 2, asks the operator to confirm and make a selection, determines the corrected character string,
It is sent to the character string correction section 11.

The character display control section 5 displays information input from the character reading section, the abnormal character string detection section 16, and the automatic correction string determination section, and requests the operator to confirm, select, and input a correction string. It also displays character strings that are input multiple times through the modified character string input section 7.

The modified character string input section inputs the operator's instructions and the modified character string through the character input device 3, and inputs the character display control section 5, the automatic modified character string determination section 10. The character string correction unit 111 is sent.

8- The character string correction unit 11 performs corrections input from the automatically corrected character string determining unit 10 and the corrected character string input unit 7 to the text that is input in large numbers from the character reading unit 12, and stores the corrected text in the text storage device 13. Store again.

Next, the text reading function will be explained.

That is, the pronunciation analysis unit 14 converts the text input from the text reading unit 12 into kanji, kana, etc. using the reading information input from the dictionary output unit 17 and the text analysis result input from the text analysis result storage unit 15. The mixed sentences are converted into information regarding phonemes and prosody, and the results are sent to the arbitrary word speech parameter synthesis section 8. The arbitrary word speech parameter synthesis section 8 converts multiple input information and arbitrary word speech to the reading analysis section 14. Using the acoustic parameters, prosody control information, etc. corresponding to each phoneme stored in the parameter dictionary 9, an acoustic parameter time series corresponding to the sentence is generated. and,
Speech synthesis device 4 The acoustic parameters inputted from the arbitrary language speech parameter synthesis section are converted into speech waves by a speech synthesis module (which may be composed of an LSI, etc.), and then sent to the operator through the speech output device 1. It can be output.

Next, the embodiment of the present invention shown in FIG. 2 will be explained.

In FIG. 2 1, 21 is a display section, 22 is an input section, 23
Key control unit, 26-1 to 26-n are character string unit dictionaries, 2
5 dictionary selection section, 27 is a text file, 28 is a text reading section, 29 is a collation section, 30 is a reading modification section, 31 is a speech synthesis section, and 32 is a speech output section.

Each of the character string unit dictionaries 26-1 to 26-n stores character strings that may be used in a sentence along with their pronunciations, and which character string is stored in which character string unit dictionary is determined. , determined by the type of character string (word, phrase, etc.). In other words, there are two dictionaries for each type of string.
6-1 to 26-n are determined.

The dictionary selection section may be a simple switch mechanism as shown in the illustration, and depending on the type of text, the necessary character string dictionary 26-1 can be selected.
26-n is selected and its contents are sent to the matching section 29.

The text read out from the text file 27 by the text reading section 28 is checked by the matching section 29 in character string unit dictionaries 26-1 to 26-n selected by the dictionary selection section 25 in character string units such as words. When a word in a string is fixed, how to read it from string unit dictionaries 26-1 to 26-n
Accordingly, the reading is converted by the reading converter 30, the speech is synthesized by the speech synthesizer 31, and the speech is outputted by the speech output section 32), so that the sentence is read out.

The main control unit 23 controls the operations described above, and further sets priorities for the character string unit dictionaries 26-1 to 26-nl.
Dictionary according to this priority? The switch of the folding section 25 is controlled, and the dictionary selection section 25 is provided with character string unit dictionaries 26-1 to 26-.
It is configured to perform priority selection control of n.

(Q: As described in detail, according to the present invention, character string unit dictionaries are constructed for each type of character string, and are selectively used depending on the type of text. This improves collation efficiency and improves usage. It is economical because it can provide a character string unit dictionary for each walking distance that suits the individuality of the person.

11-

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of a text inspection and correction device proposed by the inventors to which the text reading device of the present invention is applied, and FIG. 2 is a diagram showing an embodiment of the present invention. Next, the embodiment of the present invention shown in FIG. 2 will be explained. In FIG. 2, 21 is a display section, 22 is an input section, 23 is a main control section, 26-1 to 26-n are character string unit dictionaries, and 25
27 is a dictionary selection section, 27 is a text file, 28 is a text reading section, 29 is a digit matching section, 30 is a reading conversion section, 31 is a speech synthesis section, and 32 is a speech output section. 12-

Claims (1)

[Claims] (1) In a text reading inspection device that has a dictionary for each character string that stores information for each character string that may be used in a sentence, and that reads the text aloud while comparing it with character strings in the text. , comprising a plurality of character string unit dictionaries, and a selection means for selecting the plurality of character string unit dictionaries, and each character string unit dictionary stores a different type of character string, and the character string is selected according to the type of the sentence. A text reading inspection device characterized in that the plurality of character string unit dictionaries are selectively used by a selection means, (2) the selection means further assigning priority to the plurality of character string unit dictionaries; Claim 1 characterized by selective use
A reading inspection device for sentences written in the section.