Title:
METHOD FOR CONVERTING VOICE FEATURE OF VOICE
Document Type and Number:
WIPO Patent Application WO/2022/108040
Kind Code:
A1
Abstract:
A method for converting the voice feature of a voice according to an embodiment of the present invention can comprise the steps of: generating a first audio vector corresponding to a first voice by means of a first artificial neural network, the first audio vector comprising a text feature value of the first voice, a voice feature value of the first voice and a style feature value of the first voice in an indistinguishable manner, and the first voice being an utterance of a first text by a first speaker; generating a first text feature value corresponding to the first text by means of a second artificial neural network; generating a second audio vector, which is the first audio vector having the voice feature value of the first voice removed therefrom, by means of the first text feature value and a third artificial neural network; and generating a second voice to which the feature of a target voice is applied, by means of the second audio vector and a voice feature value of the target voice.
More Like This:
Inventors:
CHOI HONG SEOP (KR)
PARK SEUNG WON (KR)
PARK SEUNG WON (KR)
Application Number:
PCT/KR2021/010116
Publication Date:
May 27, 2022
Filing Date:
August 03, 2021
Export Citation:
Assignee:
MINDS LAB INC (KR)
International Classes:
G10L13/033; G06N3/08; G10L13/08; G10L15/04; G10L15/06; G10L15/26
Foreign References:
| KR101666930B1 | 2016-10-24 | |||
| US20180342256A1 | 2018-11-29 | |||
| JP2008058696A | 2008-03-13 |
Other References:
DESAI, SRINIVAS ET AL.: "Voice conversion using Artificial Neural Networks", 2009 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, 19 April 2009 (2009-04-19), pages 3893 - 3896, XP031460124, Retrieved from the Internet
AYODEJI AGBOLADE OLAIDE; OYETUNJI S.A: "Voice conversion using coefficient mapping and neural network", 2016 INTERNATIONAL CONFERENCE FOR STUDENTS ON APPLIED ENGINEERING (ISCAE), SCHOOL OF MECHANICAL AND SYSTEMS ENGINEERING, 20 October 2016 (2016-10-20), pages 479 - 483, XP033038984, DOI: 10.1109/ICSAE.2016.7810239
KOTANI, GAKU ET AL.: "Voice conversion based on deep neural networks for time-variant linear transformations", 2017 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC, 12 December 2017 (2017-12-12), pages 1259 - 1262, XP033315600, Retrieved from the Internet DOI: 10.1109/APSIPA.2017.8282216
AYODEJI AGBOLADE OLAIDE; OYETUNJI S.A: "Voice conversion using coefficient mapping and neural network", 2016 INTERNATIONAL CONFERENCE FOR STUDENTS ON APPLIED ENGINEERING (ISCAE), SCHOOL OF MECHANICAL AND SYSTEMS ENGINEERING, 20 October 2016 (2016-10-20), pages 479 - 483, XP033038984, DOI: 10.1109/ICSAE.2016.7810239
KOTANI, GAKU ET AL.: "Voice conversion based on deep neural networks for time-variant linear transformations", 2017 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA ASC, 12 December 2017 (2017-12-12), pages 1259 - 1262, XP033315600, Retrieved from the Internet
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF:
Previous Patent: FORMWORK PANEL
Next Patent: TANDEM BIO-ORGANIC LIGHT-EMITTING DIODE FOR PHOTODYNAMIC THERAPY AND PHOTODYNAMIC APPARATUS COMPRISI...
Next Patent: TANDEM BIO-ORGANIC LIGHT-EMITTING DIODE FOR PHOTODYNAMIC THERAPY AND PHOTODYNAMIC APPARATUS COMPRISI...