JPS59127146A - Sentence reading-out device - Google Patents

Abstract

PURPOSE:To obtain a sentence reading device which ensures the proper pronunciation by using a dictionary in accordance with the special properties of words such as technical terms that require special pronunciation. CONSTITUTION:A sentence read out of a sentence file 27 through a sentence reading part 28 is collated with a dictionary 26 every word and then converted by a reading converting part 30 in accordance with the reading way obtained from the dictionary 26 when identification of words is obtained. Then the sounds are synthesized at a rule synthesizing part 31, and the sounds are delivered through a sound output part 32. While the writing and reading ways of words such as technical terms, etc. that require special ways of reading are registered and recorded to the dictionary 26 via an input part 22. If a special word (e.g., VTAM) is contained in a sentence, a sound data on the special reading ''BUITAMU'' is delivered.

Description

[Detailed Description of the Invention] Technical Field of the Invention The present invention relates to a text processing system that automatically detects and corrects typos and omitted characters in text in language processing used in printing, publishing, and other business operations. This relates to a text reading device that can be applied to abnormal practice correction devices, etc. ■) Many parts of operations such as purchasing, printing, and publishing temporary techniques 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 is most likely a typographical error due to a misunderstanding on the part of the ministry, or a typo or omission during input. In addition, when publishing newspapers, etc., there are limits to the range of usage that can be used, contrary to social common sense, and at present it is essential that humans inspect and proofread texts input into machines. . This manual inspection and calibration requires a great deal of effort and time. In addition, manual proofreading also has errors and omissions, and double or triple inspection and proofreading may be required in some cases. In order to carry out such inspection and proofreading, automatic text inspection and correction devices have been proposed. However, in languages such as Japanese, where word boundaries are unclear, the above method could not be used, so the inventors A text abnormality inspection and correction device suitable for language processing on Japanese ships, etc. has been proposed. It is desirable to have a device that not only displays the verification results, correction methods, etc. on the display, but also converts them into audio and reads them out.The device is a device that inputs the text manuscript into a format that can be handled by the information device. It is also used when comparing a document with a character code string entered in a word processor, etc. By the way, conventional document reading devices could only output audio using the readings stored in a dictionary. However, depending on the purpose of use, Words that are easily misunderstood or that are confused with other words may be pronounced differently from the normal reading.This tendency is especially strong for industry terms.Therefore, such words must be read in a special way depending on each word. It is necessary to output audio.

(6) Object of the Invention Therefore, as described above, an object of the present invention is to provide a text reading device capable of outputting a text in a special reading method different from the normal reading method.

(G) Structure of the Invention In order to achieve this object, the text reading device of the present invention has a dictionary that stores the notations and pronunciations of words, clauses, phrases, etc., and matches the written sentences with the dictionary. A text reading device that identifies words, etc., converts them into pronunciations, and outputs them aloud, is characterized by having a means for inputting the pronunciations in the dictionary, making it possible to register the notation and/or pronunciation of words, etc. .

■ 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 FIG. 1, 1 is a speaker, an audio output device for headphones, 2 is a character display device such as Japanese Disz 1/I, etc.
3 is a character input device such as a kanji input keyboard, 4 is a speech synthesizer, 5 is a character display control unit, 3-6 is a system control unit that controls the entire system, 7
8 is a modified character string input unit, 8 is an arbitrary language speech parameter synthesis unit, 9 is an arbitrary language speech parameter dictionary, 10 is an automatically corrected character string determination unit, 11 is a character string correction unit, 12 is a text reading unit, and 13 is a text 14 is a reading analysis unit; 15 is a sentence analysis result storage unit; 16 is an abnormal character string detection unit; 17 is a dictionary reading unit; 12 from the text storage device 13 and sent to the abnormal character string detection section 16 and character display control section 5. The text storage device 13 is a magnetic disk, a semiconductor storage device, etc., and stores texts (character code strings) as well as text numbers, page numbers, and text numbers necessary for managing texts.
Memorizes line numbers, etc. The text reading unit 12 sequentially reads out the target text for inspection from the text storage device 13 in units of sentences or small units delimited by punctuation marks, commas, parentheses, etc.
send to The text input to the abnormal character string detection section is compared with the character string read out from the dictionary 18 by the dictionary output section 17.

Character strings/yomiji 4118 contains character strings (words, compound candy) that may appear in sentences, which cannot be used for automatic correction, and their connection relationship information (combinations that are allowed in sentences, Discrimination of combinations that are not allowed), character strings that are unlikely to appear in sentences (incorrect words), how to correct them (correct word strings, etc.), and back door information for use in reading (reading, word accent type). One group or multiple groups are memorized. The dictionary extraction section 17 is equipped with an address table, etc., for quickly reading out necessary information from the character string/reading dictionary 18, and reads out character strings in a sentence according to instructions from the abnormal character string detection section 16 or the reading analysis section 14. An abnormal character detection unit 1 reads out information regarding a sentence string in the dictionary that matches from the dictionary 18 and sends it to the abnormal character string detection unit 16 or reading analysis unit 14 or automatic correction character string determination unit 10.
In step 6, it is determined whether the combination of the character string of the test target sentence read out from the sentence storage device 13 and the character string read out from the dictionary by the output unit 17 matches, and whether the combination is allowed. or the string that matches the incorrect word.
It is determined whether or not it exists in the text to be inspected, and the result is sent to the automatically corrected character string determination section 10 and the 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 combination exists can be applied, and any method can be easily realized by a person skilled in the art. In addition, checking whether the above combination is allowed is as follows:
For each character string in the dictionary, its forward connection characteristics and backward connection characteristics are memorized, and the allowed combinations of backward connection characteristics and forward connection characteristics are also stored separately. In this case, the backward connection characteristics 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 result of the sentence analysis (information such as how to combine dictionary character strings that match the sentence) to be used in the reading analysis unit 14, if necessary. , memorize. As a result of the above, 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 as an abnormality of the sentence, and then Send to 5. The automatic correction string determination section is the abnormal string detection section 1.
Wrong words input from 6 and dictionary output section 17
Based on the correction method inputted by the operator, if necessary, the correction method is sent to the character display control unit 5, displayed on the character display device 2, etc., the operator is asked to confirm and select, and the corrected character string is displayed. It is determined and sent to the character string correction section 11. Character display control section 5
Then, information input from the character reading section, the abnormal character string detection section 16, and the automatically corrected character string determination section is displayed, and the operator is asked to confirm, select, and input a corrected string. It also displays the character string input from the modified character string input section 7. The modified character string input section inputs the operator's instructions and the modified character string from the character input device 3, and inputs the character display control section 5, the automatic modified character string determination section 10. It is sent to the character string correction unit 11. Character string modification 7 - The pressure section 11 corrects text that is input in large numbers from the character reading section 12.
The corrections input from the automatically corrected character string determination unit 10 and the corrected character string input unit 7 are performed, and the corrected sentences are stored in the text storage device 1.
In 3rd, he performs the performance again.

Next, the text reading device will be explained. That is,
The pronunciation analysis unit 14 extracts the text input from the text reading unit 12 from the dictionary and uses the pronunciation information input to the unit 17 and the text analysis results input from the text analysis result storage unit 15 to extract kanji, kana, etc. The mixed sentences are converted in terms of phonemes and prosody, and the results are sent to the arbitrary word speech parameter synthesis unit 8.
The arbitrary word speech parameter synthesis section 8 uses the information input from the pronunciation analysis section 14 and the acoustic parameters and prosodic control information corresponding to each phoneme stored in the arbitrary word speech parameter dictionary 9. Then, an acoustic parameter time series corresponding to the sentence is generated. Then, the speech synthesis device 4 converts the input acoustic parameters into speech waves using a speech synthesis module (which may be composed of an LSI or the like), and outputs them through the speech output device 1. Operation 8 - It is possible to output to the person.

Next, the embodiment of the present invention shown in FIG. 2 will be explained. In the second,
21 is a display section, 22 is an input section, 23 is a main control section, 24 is a dictionary registration section, 26 is a dictionary, 27 is a text file, 28 is a text reading section, 29 is a matching word identification section, 30 is a reading conversion section , 31 is a rule synthesis section, and 32 is an audio output section. The text read out from the text file 27 by the text reading unit 28 is matched word by word (or phrase by phrase, etc.) with the dictionary 26, and when a word is identified, the word is '!' The text is converted by the pronunciation converting section 30 according to the pronunciation from 26, synthesized into speech by the rule synthesis section 31, and outputted by the speech output section 32, so that the sentence is read out. In the dictionary 26, the notation and/or pronunciation of words that are read in a special way (for example, industrial candy, etc.) are registered and recorded via the input section 22, and when the word exists in one sentence, the corresponding word is recorded. The pronunciation from the dictionary 26 is outputted as voice in the same way as in the previous example ◇ Therefore, it becomes possible to read the text in the reading method desired by the author of the text 〇 (Q As described in detail, the features of the present invention provides a means to input readings such as morning paper→asakan, allows special readings to be registered in the dictionary, and also allows for the notation and reading of abbreviations that are not recognized in the dictionary, such as VTAM→buitam, AIM→niimu, etc. It is also possible to register the text in the dictionary, and the author can output the pronunciation as desired.

[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. In FIG. 2, 21 is a display section, 22 is an input section, 23 is a main control section, and 24 is a dictionary registration section. 26 is a dictionary, 27 is a text file, 28 is a text reading unit, 29 is a matching word identification unit, 30 is a reading conversion unit, 3
1 is a rule synthesis section, and 32 is an audio output section. 11-

Claims (1)

[Claims] In a text reading device that has a dictionary that stores the notations and pronunciations of vulgar words, clauses, phrases, etc., identifies words, etc. by matching the written sentences with the dictionary, converts them into pronunciations, and outputs them aloud. A text reading device 0 characterized by having a means for inputting the pronunciation in a dictionary and being able to register the notation and/or pronunciation of words, etc.