METHOD FOR GENERATING PRONUNCIATION DICTIONARY AND APPARATUS THEREFOR

Title:

METHOD FOR GENERATING PRONUNCIATION DICTIONARY AND APPARATUS THEREFOR

Document Type and Number:

WIPO Patent Application WO/2019/208859

Kind Code:

A1

Abstract:

Disclosed are a method for generating a pronunciation dictionary and an apparatus therefor. Specifically, a method for generating a pronunciation dictionary by a voice recognition device may comprise the steps of: receiving an input of a voice from a user or a machine; recognizing the input voice by using an audio model; generating a pronunciation symbol for the voice on the basis of a result of the recognition; receiving an input of a user vocabulary from the user; and mapping the input user vocabulary to the generated pronunciation symbol, and storing same in the pronunciation dictionary.

More Like This:

WO/2013/122909	REAL TIME CLOSED CAPTIONING LANGUAGE TRANSLATION
WO/2017/072307	METHOD AND APPARATUS FOR IMPROVING RECOGNITION ACCURACY FOR THE HANDWRITTEN INPUT OF ALPHANUMERIC CHARACTERS AND GESTURES
WO/2022/037388	VOICE GENERATION METHOD AND APPARATUS, DEVICE, AND COMPUTER READABLE MEDIUM

Inventors:

HWANG MYEONGJIN (KR)
JI CHANGJIN (KR)

Application Number:

PCT/KR2018/004971

Publication Date:

October 31, 2019

Filing Date:

April 27, 2018

Export Citation:

Click for automatic bibliography generation Help

Assignee:

SYSTRAN INT (KR)

International Classes:

G10L15/26; G06F16/00; G10L15/01; G10L15/18

Foreign References:

US20020013707A1	2002-01-31
KR20160098910A	2016-08-19
KR20100130263A	2010-12-13
KR100277694B1	2001-01-15
KR20130011323A	2013-01-30

Attorney, Agent or Firm:

JANG, Youngjin (KR)

Download PDF:

View/Download PDF PDF Help

Previous Patent: VOICE RECOGNITION METHOD AND DEVICE THEREFOR

Next Patent: METHOD FOR RECORDING AND OUTPUTTING CONVERSATION BETWEEN MULTIPLE PARTIES USING VOICE RECOGNITION TE...