シェア

知能コミュニケーション研究室のQuoc Truong Doさん(博士後期課程2年)が、IEEE Signal Processing Society Japan Student Journal Paper Awardを受賞しました。
Quoc Truong Do received the IEEE Signal Processing Society Japan Student Journal Paper Award. (2017/11/14)

　 IEEE Signal Processing Society(SPS)の Tokyo Joint Chapter, Kansai Chapter, Sendai Chapterでは、次世代を担う学生の研究を奨励するため、Student Paper Awardを設け、優れた英文論文を発表した学生を表彰しています。 tanaka
  • 受賞者 Awardee:
    Quoc Truong Do
  • 著者 Authors:
    Quoc Truong Do, Tomoki Toda, Graham Neubig, Sakriani Sakti, Satoshi Nakamura
  • 受賞研究テーマ
    "Preserving word-level emphasis in speech-to-speech　translation "
    　Speech-to-speech translation (S2ST) is a technology that translates speech across languages, which can remove barriers in cross-lingual communication. In conventional S2ST systems, the linguistic meaning of speech was translated, but paralinguistic information conveying other features of the speech such as emotion or emphasis were ignored. In this paper, we propose a method to translate paralinguistic information, specifically focusing on emphasis. The method consists of a series of components that can accurately translate emphasis using all acoustic features of speech. First, linear-regression hidden semi-Markov models (LR-HSMMs) are used to estimate a real-numbered emphasis value for every word in an utterance, resulting in a sequence of values for the utterance. After that, the emphasis translation module translates the estimated emphasis sequence into a target language emphasis sequence using a conditional random field (CRF) model considering the features of emphasis levels, words, and part-of-speech tags. Finally, the speech synthesis module synthesizes emphasized speech with LRHSMMs, taking into account the translated emphasis sequence and transcription. The results indicate that our translation model can translate emphasis information, correctly emphasizing words in the target language with 91.6% F-measure by objective evaluation. A listening test with human subjects further showed that they could identify the emphasized words with 87.8% Fmeasure, and that the naturalness of the audio was preserved

  • 受賞者のコメント Awardee's voice
    　It is an honor to receive this award. I am very happy that my work is recognized by the IEEE SPS Japan and I would like to give my sincere gratitude to all of my co-authors for valuable comments and discussion. We will continue the research to produce more and more interesting results for the research community.

  • 外部リンク Links to：
    IEEE SPS Tokyo Joint Chapter HP:https://ieee-jp.org/section/tokyo/chapter/SP-01/sp.htm

>> 知能コミュニケーション研究室 Augmented Human Communication lab.

CLOSE