Wed, Dec 1 AM 10:15 - 10:30 |
|
10:15-10:30 |
( 15 min. ) |
Wed, Dec 1 AM 10:30 - 12:00 |
(1) |
10:30-11:00 |
|
(2) |
11:00-11:30 |
|
(3) |
11:30-12:00 |
|
|
12:00-13:00 |
( 60 min. ) |
Wed, Dec 1 PM 13:00 - 14:00 |
(4) |
13:00-14:00 |
|
|
14:00-14:20 |
( 20 min. ) |
Wed, Dec 1 PM 14:20 - 15:40 |
(5) |
14:20-14:40 |
|
(6) |
14:40-15:10 |
|
(7) |
15:10-15:40 |
|
|
15:40-16:00 |
( 20 min. ) |
Wed, Dec 1 PM 16:00 - 17:30 |
(8) |
16:00-16:30 |
|
(9) |
16:30-17:00 |
|
(10) NLC |
17:00-17:30 |
Digital Evolution of Life Gave Birth to Language
-- Law of Digital Language -- |
Kumon Tokumaru (Writer) |
Thu, Dec 2 AM 10:30 - 12:00 |
(11) |
10:30-11:00 |
|
(12) |
11:00-11:30 |
|
(13) SP |
11:30-12:00 |
Multi-faceted assessment of language learners' ability of perception and production of English speech based on shadowing |
Takuya Kunihara, Chuanbo Zhu, Daisuke Saito, Nobuaki Minematsu (UTokyo), Noriko Nakanishi (KGU) |
|
12:00-13:00 |
( 60 min. ) |
Thu, Dec 2 PM 13:00 - 14:00 |
(14) |
13:00-14:00 |
|
|
14:00-14:20 |
( 20 min. ) |
Thu, Dec 2 PM 14:20 - 16:20 |
(15) SP |
14:20-14:50 |
End-to-End Speech Recognition System Using Sparse Representation |
Reiichiro Yasaki, Makoto Ohki (Yamanashi Univ.) |
(16) SP |
14:50-15:20 |
Music Separation by Regularized NMF using CQCC |
Kohei Miyajima, Makoto Ohki (Yamanashi Univ.) |
(17) SP |
15:20-15:50 |
improvement of multilingual speech emotion recognition by normalizing features using CRNN |
Jinhai Qi, Motoyuki Suzuki (OIT) |
(18) SP |
15:50-16:20 |
Simultaneous measurement of linear, non-linear, random responses of pitch extrctors to frequency modulated voiced sounds
-- application of extended time-stretched-pulse sequence with orthogonalization -- |
Hideki Kawahara (Wakayama Univ.), Ken-Ichi Sakakibara (Health Science Univ. okkaido), Kohei Yatabe (Waseda Univ.), Tatsuya Kitamura (Konan Univ.), Hideki Banno (Meijo Univ.), Masanori Morise (Meiji Univ.) |
|
16:20-16:40 |
( 20 min. ) |
Thu, Dec 2 PM 16:40 - 18:10 |
(19) |
16:40-17:10 |
|
(20) |
17:10-17:40 |
|
(21) NLC |
17:40-18:10 |
Adding span loss to BERT for opinion target extraction |
Takeshi S. Kobayakawa (NHK) |
|
- |
|
Fri, Dec 3 AM 10:30 - 12:00 |
(22) SP |
10:30-11:00 |
An approach to voice conversion for manipulating emotion dimensions |
Keita Mukada, Hiroki Mori (Utsunomiya Univ.) |
(23) SP |
11:00-11:30 |
Multi-speaker Audiobook Speech Synthesis using Discrete Character Acting Styles Acquired by VQVAE |
Wataru Nakata, Tomoki Koriyama, Shinnosuke Takamichi, Yuki Saito (UT), Yusuke Ijima, Ryo Masumura (NTT), Hiroshi Saruwatari (UT) |
(24) |
11:30-12:00 |
|
|
12:00-13:00 |
( 60 min. ) |
Fri, Dec 3 PM 13:00 - 14:20 |
(25) |
13:00-14:20 |
|
|
14:20-14:40 |
( 20 min. ) |
Fri, Dec 3 PM 14:40 - 16:40 |
(26) |
14:40-15:10 |
|
(27) |
15:10-15:40 |
|
(28) |
15:40-16:10 |
|
(29) NLC |
16:10-16:40 |
Label smoothing with co-occurrences information for multi-label classification |
Yuki Yasuda, Taichi Ishiwatari, Taro Miyazaki, Jun Goto (NHK) |
Fri, Dec 3 PM 16:40 - 17:00 |
|
16:40-17:00 |
( 20 min. ) |