Title:
SPEAKER EMBEDDING APPARATUS AND METHOD
Document Type and Number:
WIPO Patent Application WO/2021/005741
Kind Code:
A1
Abstract:
An input unit 81 inputs an observation at current time step. A frame alignment unit 82 computes a frame alignment at a current time step by using the input observation. An i-vector computation unit 83 computes an i-vector and a precision matrix by using the computed frame alignment, the input observation, and a product obtained when computing the i-vector at the previous time step. An output unit 84 outputs the computed i-vector and precision matrix.
Inventors:
LEE KONG AIK (JP)
KOSHINAKA TAKAFUMI (JP)
KOSHINAKA TAKAFUMI (JP)
Application Number:
PCT/JP2019/027304
Publication Date:
January 14, 2021
Filing Date:
July 10, 2019
Export Citation:
Assignee:
NEC CORP (JP)
International Classes:
G10L17/18; G10L17/02; G10L17/04
Foreign References:
| US20190005962A1 | 2019-01-03 |
Other References:
GLEMBEK, 0. ET AL.: "SIMPLIFICATION AND OPTIMIZATION OF I-VECTOR EXTRACTION", PROC. ICASSP 2011, May 2011 (2011-05-01), pages 4516 - 4519, XP032001683, DOI: 10.1109/ICASSP.2011.5947358
See also references of EP 3997696A4
See also references of EP 3997696A4
Attorney, Agent or Firm:
IWAKABE Fuyuki et al. (JP)
Download PDF: