次世代ネットワークにおける超臨場感音響相互通信の実現 -Ultra-realistic acoustic interactive communication on next-generation Internet-

Seminar Program

  • The 9th seminar of A3 Foresight Program
  • "Ultra-realistic Acoustic Interactive Communication on Next-generation Internet"
    October 31, 2015, Beijing Institute of Technology, China


    image025 class=PopBoxImageSmall image025 class=PopBoxImageSmall

    Date: October 31, 2015
    Seminar Venue: Beijing Institute of Technology, China


    09:00-09:05 Welcome remark by Yonghong Yan
    09:05-09:10 Taking Group photo

    Session 1 chairpersons: Xiang Xie, Masashi Unoki

    • - 09:10-09:30 Xu Huaxing (Institute of Acoustics, China) An Stochastic Robust Approximation Method with Enhanced Robustness for Crosstalk
    • - 09:30-09:50 Kang Hyun Lee (Seoul National University, Korea) Training techniques for environment-robust DNN-HMM
    • - 09:50-10:10 Liqiang Xu (Beijing Institute of Technology, China) A Telephone Customer Service Speech Database of Call Center
    • - 10:10-10:30 Xingfeng LI (JAIST, Japan) Toward Improving Estimation Accuracy of Emotion Dimensions in Bilingual Scenario Based on Three-layered Model

    10:30-10:50 Coffee break
    Session 2 chairpersons: Dongwen Ying, Jong Won Shin
    • - 10:50-11:10 Isao Miyagawa (Tohoku University, Japan ) Singing Error Detection from the Singing Voice based on Dynamic Time Warping
    • - 11:10-11:30 Inkyu Choi (Seoul National University, Korea) Detection and Classification of Polyphonic Acoustic Events
    • - 11:30-11:50 Jing Liu (Beijing Institute of Technology, China) Dialogues Satisfaction Analysis Based on Emotion Recognition in Call Centers
    • - 11:50-12:10 Yang Liu (JAIST, Japan) RESTORATION FOR INSTANTANEOUS AMPLITUDE AND PHASE OF SPEECH SIGNAL USING KALMAN FILTER IN NOISY REVERBERANT ENVIRONMENTS

    12:10-13:30 Lunch time
    Session 3 chairpersons: Junfeng Li, Shuichi Sakamoto
    • - 13:30-13:50 Taketo Kase (Tohoku University, Japan) Development of Spoken Dialogue System Generating Emotional Speech Response Based on Utterance State Estimation
    • - 13:50-14:10 Song-Kyu Park (Hanyang University, Korea ) GSC incorporating Data-Driven Post-Filtering for Speech Recognition
    • - 14:10-14:30 Fang qianqian (Institute of Acoustics, China) Equalization of sound reproduction system based on the human perception characteristics
    • - 14:30-14:50 Cesar SALVADOR (RIEC. Japan) Distance Extrapolation of Head-Related Transfer Functions on the Horizontal Plane
    • - 14:50-15:10 Xu Li (Beijing Institute of Technology, China) Combining non-negative sparse coding and factorial conditional random field for single-channel speech separation
    • - 15:10-15:30 Emika Takeishi (Tohoku University, Japan) Design and evaluation of phonetically and prosodically balanced emotion-dependent sentence set for emotional speech database

    15:30-15:50 Coffee break
    Session 4 chairpersons: Jing Wang, Jorge TREVINO
    • - 15:50-16:10 Ziteng Wang (Institute of Acoustics, China) Incorporating DNN based speech enhancement into speech recognition
    • - 16:10-16:30 Yingnan Zhang (Beijing Institute of Technology, China) A Survey for Social Personality Analysis Based on Prosodic Feature
    • - 16:30-16:50 Yuki Saito (Tohoku University, Japan ) Facial image conversion based on transformation of Animation Units using a DNN
    • - 16:50-17:10 Subin Lee (Seoul National University, Korea) Exploring acoustic markers from speech for diagnosis of depression

    17:10-17:20 Closing ceremony
    17:20- Dinner time
    19:30- PI meeting