VOICE SYNTHESIS APPARATUS WHICH PROCESSES SPACING ON READING FOR SENTENCES AND THE OPERATING METHOD THEREOF
Disclosed are a speech synthesis device that performs spacing-reading processing for a sentence and an operating method thereof. The present invention relates to a technology that, based on an average pronunciation required time for each phoneme acquired and an average silent syllable duration time...
Gespeichert in:
Hauptverfasser: | , , , |
---|---|
Format: | Patent |
Sprache: | eng ; kor |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Disclosed are a speech synthesis device that performs spacing-reading processing for a sentence and an operating method thereof. The present invention relates to a technology that, based on an average pronunciation required time for each phoneme acquired and an average silent syllable duration time and appearance ratio of silent syllable per word segment from the predetermined training sentences, supports a generation of a more natural synthetic speech, when the speech synthesis for the predetermined sentence is performed, by inserting a silent syllable section at the end of the selected word after selecting the word predicted to have the silent syllable section among the words existing in the sentence. The speech synthesis device comprises: a pronunciation required time information storage part; a silent syllable time information storage part; a ratio storage part; an information extraction part; a designation part; a word selection part; a silent syllable mark inserting part; and a speech synthesis part.
문장에 대한 띄어읽기 처리를 수행하는 음성 합성 장치 및 그 동작 방법이 개시된다. 본 발명은 소정의 훈련 문장들로부터 획득된 음소별 평균 발음 소요 시간과 평균 묵음 지속 시간 및 어절별 묵음의 등장 비율을 기초로, 소정의 문장에 대한 음성 합성이 수행될 때, 상기 문장에 존재하는 어절들 중 묵음 구간을 갖는 것으로 예측되는 어절을 선택한 후 선택된 어절 후단에 묵음 구간을 삽입함으로써, 보다 자연스러운 합성 음성이 생성될 수 있도록 지원하는 기술에 대한 것이다. |
---|