KR930010711A - How to Read Sentences for Document Speech Converters - Google Patents

How to Read Sentences for Document Speech Converters Download PDF

Info

Publication number
KR930010711A
KR930010711A KR1019910021272A KR910021272A KR930010711A KR 930010711 A KR930010711 A KR 930010711A KR 1019910021272 A KR1019910021272 A KR 1019910021272A KR 910021272 A KR910021272 A KR 910021272A KR 930010711 A KR930010711 A KR 930010711A
Authority
KR
South Korea
Prior art keywords
word
read
investigation
mother
possessive
Prior art date
Application number
KR1019910021272A
Other languages
Korean (ko)
Inventor
하남규
Original Assignee
정용문
삼성전자 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 정용문, 삼성전자 주식회사 filed Critical 정용문
Priority to KR1019910021272A priority Critical patent/KR930010711A/en
Publication of KR930010711A publication Critical patent/KR930010711A/en

Links

Abstract

문장음성 변환방법에 관한 것으로, 특히 읽어들인 문장의 적절한 위치에 끊어읽기에 관한 정보를 삽입하는 방법에 관한 것이다.The present invention relates to a sentence-to-speech conversion method, and more particularly, to a method of inserting information about reading at an appropriate position of a read sentence.

문장을 어절단위로 순차 읽어들이는 제1단계와, 상기 제1단계에서 읽어 들인 임의의 어절 Mi가 어미이고 그 다음 어절 Mi+1은 어미가 아닐경우 상기 M다음에 쉬어읽기 기호를 삽입하는 제2단계와, 상기 제1단계에서 임의의 두어절 상기 Mi와 상기 Mi+1이 모두 어미일때 읽어들인 어절이 연속발음 규칙이 아니면 상기 Mi다음에 상기 쉬어읽기 기호를 삽입하고 연속발음 규칙일 경우는 상기 Mi와 Mi+1를 붙이는 제3단계와, 상기 제3단계 수행후 상기 i가 임의의 n값보다 작을경우 상기 제2단계로 리턴하고 i가 n값보다 클경우는 종료하는 제4단계와, 상기 제2단계에서 만일 상기 Mi가 어미가 아닐경우 상기 Mi가 주격조사인가 점검하여 주격조사이면 상기 Mi다음에 쉬어읽기 부호를 삽입한후 상기 제4단계로 리턴하는 제5단계와, 상기 제5단계에서 상기 Mi가 주격조사가 아니고 소유격조사이면 바로 앞어절이 수식언인가 점검하여 수식언 일시 상기 Mi다음에 끊어읽기 부호를 부여하고 앞어절이 수식언이 아닐경우는 상기 Mi+1과 붙이는 제6단계와, 상기 제6단계에서 상기 Mi가 소유격조사도 아닐경우 상기 Mi가 조사인가 점검하여 조사이면 다시 상기 Mi+1과 붙이고 상기 Mi+1이 어미가 아닐경우는 상기 Mi다음에 제2끊어읽기 부호를 삽입하는 제7단계와, 상기 제7단계에서 Mi가 조사도 아닐경우는 상기 Mi+1과 붙이는 제8단계로 구성한다.A first step of sequentially reading sentences in word units; and if the word Mi read in the first step is a mother and the next word Mi + 1 is not a mother, inserting a read symbol after the M. If the word read when the word Mi and Mi + 1 are both ending words in step 2 and the first step is not a continuous phonetic rule, the word is inserted after the Mi and the continuous phonetic rule is A third step of attaching the Mi and Mi + 1, and a fourth step of returning to the second step if i is smaller than a value of n after performing the third step, and ending if i is greater than n. In the second step, if the Mi is not the mother, check whether the Mi is the main investigation, and if the main investigation, insert a rest read code after the Mi and return to the fourth step; In step 5, the Mi In case of the possessive investigation, it is checked whether the preceding clause is a formula word. If the expression word is not a formula word, a break code is given. If the preceding clause is not a formula word, the sixth step is attached to Mi + 1 and the sixth step is Mi. If is not a possessive investigation, check whether Mi is an investigation, and if it is an investigation, attach it again with Mi + 1, and if Mi + 1 is not a mother, inserting a second read code after Mi; If Mi is not irradiated in the seventh step, the eighth step is attached to Mi + 1.

Description

문서 음성변환장치를 위한 문장 끊어읽기 방법How to Read Sentences for Document Speech Converters

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음Since this is an open matter, no full text was included.

제1도는 본 발명에 따른 시스템도.1 is a system diagram according to the present invention.

제2도는 본 발명에 따른 흐름도.2 is a flow chart in accordance with the present invention.

제3도는 본 발명에 따른 어절의 묶음도.3 is a bundle of words according to the present invention.

Claims (1)

문서음성변환 장치의 끊어읽기 방법에 있어서, 문장을 어절단위로 순차 읽어들이는 제1단계와, 상기 제1단계에서 읽어들인 임의의 어절 Mi가 어미이고 그다음 어절 Mi+1은 어미가 아닐경우 상기 Mi다음에 쉬어읽기 기호를 삽입하는 제2단계와, 상기 제1단계에서 임의의 두어절 상기 Mi와 상기 Mi+1이 모두 어미일때 읽어들인 어절이 연속발음 규칙이 아니면 상기 Mi다음에 상기 쉬어읽기 기호를 삽입하고 연속발음 규칙일 경우는 상기 Mi와 Mi+1를 붙이는 제3단계와, 상기 제3단계수행후 상기 i가 임의의 n값보다 작을경우 상기 제2단계로 리턴하고 i가 n값보다 클경우는 종료하는 제4단계와, 상기 제2단계에서 만일 상기 Mi가 어미가 아닐경우 상기 Mi가 주격조사인가 점검하여 주격조사이면 상기 Mi다음에 쉬어읽기 부호를 삽입한후 상기 제4단계로 리턴하는 제5단계와, 상기 제5단계에서 상기 Mi가 주격조사가 아니고 소유격조사이면 바로 앞어절이 수식언인가 점검하여 수시언 일시 상기 Mi다음에 끊어읽기 부호를 부여하고 앞어절이 수식언이 아닐경우는 상기 Mi+1과 붙이는 제6단계와, 상기 제6단계에서 상기 Mi가 소유격조사도 아닐경우 상기 Mi가 조사인가 점검하여 조사이면 다시 상기 Mi+1과 붙이고 상기 Mi+1이 어미가 아닐경우는 상기 Mi다음에 제2끊어읽기 부호를 삽입하는 제7단계와, 상기 제7단계에서 Mi가 조사도 아닐경우는 상기 Mi+1과 붙이는 제8단계로 구성함을 특징으로 하는 문서음성 변환장치를 위한 문장 끊어읽기 방법.A method of breaking a document speech converter, the first step of sequentially reading sentences in word units, and if any word Mi read in the first step is a mother and the next word Mi + 1 is not a mother, A second step of inserting a read symbol after Mi, and a word read when the word Mi and Mi + 1 are both arbitrary words in the first step, if the word read is not a continuous phonetic rule and reads the word after Mi A third step of inserting a symbol and attaching the Mi and Mi + 1 in the case of a continuous phonetic rule, and returning to the second step if i is smaller than an arbitrary value of n after performing the third step, and i is the value of n If greater than the fourth step to end, and if the Mi is not the mother in the second step check whether the Mi is the main test, if the main test is inserted after the Mi read code after the Mi step 4 5th stage to return to In the fifth step, if Mi is not a subjective test but a possessive test, it is checked whether the preceding word is a formula word. In the sixth step, and if the Mi in the sixth step is not possessive investigation, check whether the Mi is investigation, and if the investigation is attached to the Mi + 1 and if the Mi + 1 is not a mother, A seventh step of inserting a second read-out code, and an eighth step of attaching the Mi + 1 if Mi is not irradiated in the seventh step; Way. ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019910021272A 1991-11-26 1991-11-26 How to Read Sentences for Document Speech Converters KR930010711A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019910021272A KR930010711A (en) 1991-11-26 1991-11-26 How to Read Sentences for Document Speech Converters

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019910021272A KR930010711A (en) 1991-11-26 1991-11-26 How to Read Sentences for Document Speech Converters

Publications (1)

Publication Number Publication Date
KR930010711A true KR930010711A (en) 1993-06-23

Family

ID=67356569

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019910021272A KR930010711A (en) 1991-11-26 1991-11-26 How to Read Sentences for Document Speech Converters

Country Status (1)

Country Link
KR (1) KR930010711A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100292376B1 (en) * 1993-11-30 2001-06-01 윤종용 Device and method for converting sentence
KR100304654B1 (en) * 1993-11-30 2001-11-22 윤종용 Method and apparatus for analyzing korean document
KR100309276B1 (en) * 1998-10-21 2001-12-17 구자홍 Apparatus and method for checking pager messages of cassette player
KR100720175B1 (en) * 2005-11-09 2007-05-18 한국전자통신연구원 apparatus and method of phrase break prediction for synthesizing text-to-speech system

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100292376B1 (en) * 1993-11-30 2001-06-01 윤종용 Device and method for converting sentence
KR100304654B1 (en) * 1993-11-30 2001-11-22 윤종용 Method and apparatus for analyzing korean document
KR100309276B1 (en) * 1998-10-21 2001-12-17 구자홍 Apparatus and method for checking pager messages of cassette player
KR100720175B1 (en) * 2005-11-09 2007-05-18 한국전자통신연구원 apparatus and method of phrase break prediction for synthesizing text-to-speech system

Similar Documents

Publication Publication Date Title
Fillmore The position of embedding transformations in a grammar
Sang et al. Introduction to the CoNLL-2001 shared task: Clause identification
Xydas et al. The DEMOSTHeNES speech composer
Maekawa et al. Corpus of Spontaneous Japanese: design, annotation and XML representation
KR930010711A (en) How to Read Sentences for Document Speech Converters
Smrz et al. Large scale parsing of Czech
Prasad et al. Comparing test-suite based evaluation and corpus-based evaluation of a wide-coverage grammar for English
KR100202292B1 (en) Text analyzer
Bachenko et al. The contribution of parsing to prosodic phrasing in an experimental text-to-speech system
Levis Prominence in Singapore and American English: Evidence from reading aloud
KR20040055288A (en) Method for tagging for prosodic module of speech synthesizer in Korean
Stoel et al. Fataluku as a tone language
Prinsloo et al. Corpus applications for the African languages, with special reference to research, teaching, learning and software
Kimura Mismatch of stress and accent in spoken Spanish
Hussain To-sound conversion for Urdu text-to-speech system
Zelinka et al. Automatic transcription of numerals in inflectional languages
Sproat Corpus-based methods and hand-build methods
Komen The Chechen it-cleft construction
Hurskainen Solutions for handling non-concatenative processes in Bantu languages
Gaved Pronunciation and text normalisation in applied text-to-speech systems.
Fofliger Oblique Pronoun Remnants on Comparative Prepositions in West Semitic1
Mettouchi et al. Prosodic segmentation and grammatical analysis in cross-linguistic corpora
Sagisaka et al. Word identification method for Japanese text-to-speech conversion system
Lammens A lexicon-based grapheme-to-phoneme conversion system.
Das et al. Hybrid sandhi-splitter for malayalam using unicode

Legal Events

Date Code Title Description
WITN Withdrawal due to no request for examination