Title:
SPEECH SYNTHESIS APPARATUS USING ARTIFICIAL INTELLIGENCE, OPERATION METHOD OF SPEECH SYNTHESIS APPARATUS, AND COMPUTER-READABLE RECORDING MEDIUM
Document Type and Number:
WIPO Patent Application WO/2020/230924
Kind Code:
A1
Abstract:
A speech synthesis apparatus using artificial intelligence, according to one embodiment of the present invention, comprises: a memory for storing a synthesized speech model and a first ratio of words classified into a minor class from among a plurality of classes; and a processor for determining a set of class classification probabilities of the words by using the words, the first ratio, and the synthesized speech model, wherein the first ratio represents a ratio in which the words are classified into the minor class within a plurality of text, the plurality of classes include: a first class corresponding to a first speech pace; a second class corresponding to a second speech pace greater than the first speech pace; and a third class corresponding to a third speech pace greater than the second speech pace, and the minor class may be a class having the lowest number from among the first to third classes.
More Like This:
JP2003302993 | VOICE SYNTHESIZER, VOICE SYNTHESIS METHOD, PROGRAM AND STORAGE DEVICE |
JPH0981175 | VOICE RULE SYNTHESIS DEVICE |
JP3267193 | VOICE READ-OUT DEVICE |
Inventors:
CHAE JONGHOON (KR)
HAN SUNGMIN (KR)
HAN SUNGMIN (KR)
Application Number:
PCT/KR2019/005814
Publication Date:
November 19, 2020
Filing Date:
May 15, 2019
Export Citation:
Assignee:
LG ELECTRONICS INC (KR)
International Classes:
G10L13/02; G06N20/00; G10L13/08
Foreign References:
US20180336882A1 | 2018-11-22 | |||
KR20170017545A | 2017-02-15 | |||
KR20150087023A | 2015-07-29 | |||
KR100499116B1 | 2005-07-04 | |||
KR20100072962A | 2010-07-01 |
Attorney, Agent or Firm:
HAW, Yong Noke (KR)
Download PDF: