Short bio

I’m Tien-Hong, a research engineer in Speech & Spoken Language Processing. I am deeply interested in Automatic Speech Recognition (ASR) and Computer-assisted Language Learning (CALL).

Currently, I am working as a Ph.D. candidate, jointly funded by Computer Science & Information Engineering @ National Taiwan Normal University (NTNU) and The Research Center for Psychological and Educational Testing (RCPET). Previously after getting an MS degree from NTNU, I stayed there, working as an R&D engineer at the RCPET.

Most of my research interests are improving ASR systems, with a minor focus on mispronunciation detection & diagnosis in CALL. In addition, I am also exploring the related machine learning topics of spoken information retrieval, natural language processing, and deep neural networks. I am an active user and lover of open-sourced ASR toolkits (e.g., kaldi and espnet) and have bits of contribution to it.

A more detailed ver. of CV is available.


Awards & Honors

  • (2023) The Second-best System: Track 3 (Hakka Pinyin), ROCLING 2023: Shared Task on Formosa Speech Recognition Challenge 2023 (FSR-2023)
  • (2023) The Third-best System: Track 2 (Hakka Hanzi), on Formosa Speech Recognition Challenge 2023 (FSR-2023)
  • (2021) High Distinction System Student Award (The Best System): Track 3, Formosa Speech Recognition Challenge 2020 (FSR 2020)
  • (2021) The Second-best System: Open Track & Closed Track, Interspeech 2021 : Shared Task on Automatic Speech Recognition for Non-Native Children’s Speech
  • (2020) IEEE Spoken Language Processing Student Travel Grant: Interspeech 2020
  • (2020) The Second-best System: Closed Track, Interspeech 2020: Shared Task on Automatic Speech Recognition for Non-Native Children’s Speech
  • (2020) The Third-best System: Open Track, Interspeech 2020: Shared Task on Automatic Speech Recognition for Non-Native Children’s Speech
  • (2018) Best Academic System Award: 2018 Formosa Speech Recognition (FSR 2018)
  • (2018) Best Paper Candidates (Two papers): Conference on Computational Linguistics and Speech Processing (ROCLING 2018), Taiwan: Paper entitled “Investigating acoustic model combination and semi-supervised discriminative training for meeting speech recognition” and “Discriminative training of acoustic models leveraging improved neural network architecture and optimization method” (in Chinese)
  • (2017) Best Paper Award: Conference on Computational Linguistics and Speech Processing (ROCLING 2017), Taiwan: Paper entitled “Exploring query intent and neural network modeling techniques for spoken document retrieval” (in Chinese)
  • (2015) The First Runner-up: Mobile Hero Smart City Develop Competition 2015 - 200K TWD