VOICE SYNTHESIZING DEVICE

PROBLEM TO BE SOLVED: To improve the quality of synthesized voice so as to obtain a natural synthetic voice by synthesizing voice on the basis of the rhythm information extracted from the input voice. SOLUTION: A rhythm information extracting unit 25 extracts the rhythm information in a voice, which...

Full description

Saved in:
Bibliographic Details
Main Authors MIYATAKE MASANORI, IIDA MASAYUKI
Format Patent
LanguageEnglish
Published 11.11.1997
Edition6
Subjects
ACOUSTICS
MUSICAL INSTRUMENTS
PHYSICS
SPEECH ANALYSIS OR SYNTHESIS
SPEECH OR AUDIO CODING OR DECODING
SPEECH OR VOICE PROCESSING
SPEECH RECOGNITION
Online AccessGet full text

Cover

More Information
Summary:PROBLEM TO BE SOLVED: To improve the quality of synthesized voice so as to obtain a natural synthetic voice by synthesizing voice on the basis of the rhythm information extracted from the input voice. SOLUTION: A rhythm information extracting unit 25 extracts the rhythm information in a voice, which is input from a microphone 30, and houses this rhythm information in a rhythm information housing unit 26. A rhythm processing unit 28 switches a changeover unit 27 to the rhythm information housing unit 26 side on the basis of a command from a rhythm change commanding unit 29, and sends a second rhythm information, which is sent from the rhythm information housing unit 26, to a voice synthesizing unit 24 of a following stage. The voice synthesizing unit 24 smoothly connects the synthesis unit data, which are sent from a synthesis unit data selecting unit 22, to each other, and adjusts the pitch on the basis of a pitch pattern as a second rhythm information, which is sent from the rhythm processing unit 28, and outputs the synthetic voice. At this stage, position information of the point of center of gravity of vowel, which is given to both the synthesis unit data and the second rhythm information, is used so as to hold the consistency in the time direction for processing.
Bibliography:Application Number: JP19960107835