Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD FOR GENERATING SPEAKER-MARKED TEXT
Document Type and Number:
WIPO Patent Application WO/2021/256614
Kind Code:
A1
Abstract:
A method for generating speaker-marked text from voice data including voices of at least two speakers, according to an embodiment of the present invention, may comprise the steps of: generating a first piece of text by converting the voice data into text; determining a speaker of each of one or more second pieces of text obtained by dividing the first piece of text into predetermined units; and providing an editing interface that displays the one or more second pieces of text and the respective speakers.

Inventors:
WON JUNG SANG (KR)
KIM HEE YEON (KR)
LIM HEE KWAN (KR)
CHOI MOO NI (KR)
NAM SEUNG MIN (KR)
YOO TAE JOON (KR)
CHOI HONG SEOP (KR)
Application Number:
PCT/KR2020/012416
Publication Date:
December 23, 2021
Filing Date:
September 15, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
MINDS LAB INC (KR)
International Classes:
G10L15/26; G06F3/0484; G10L17/02; H04N7/15
Foreign References:
KR20190143116A2019-12-30
JP2000352995A2000-12-19
KR20200011198A2020-02-03
US20130103399A12013-04-25
KR20150024188A2015-03-06
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF: