About Me: Yi-Chang Chen

Posted on August 12, 2025 in Life. View: 5,712

portrait

YC / Yi-Chang Chen / 陳宜昌

YC currently serves as a Senior Technology Manager at MediaTek Research. He has over 7 years of research and development experience in machine learning, with a focus on language, visual, and speech AI. With 4 years of team leadership experience, he adeptly leads teams through the entire process from research to application of technologies. He has published 7 machine learning papers, including in renowned conferences such as ACL, INTERSPEECH, and ICIP. Additionally, he serves as the webmaster of YC Note and the founder of Open Vibe Developers.

KEY ACHIEVEMENTS
📌 Breeze LLMs : Led the team in developing the Breeze series of Traditional Chinese LLMs, including Breeze-7B、BreeXe-8x7B、Breeze 2, achieving around 90k downloads.
📌 TASTE Spoken LM : Co–first author. First work to propose text-aligned speech tokenization for building spoken LMs, enabling native speech capabilities while preserving the original abilities of LLMs.
📌 GFD : Co–first author. Propose Generative Fusion Decoding, a novel shallow fusion framework designed to integrate LLMs into cross-modal text recognition systems for ASR and OCR. Accepted for the paper at Findings of ACL 2025.
📌 g2pW : Introduced a novel architecture for Mandarin grapheme-to-phoneme (g2p) conversion, named g2pW, achieving SoTA performance on a public dataset. Accepted for the paper at INTERSPEECH 2022.

EXPERIENCE
📌 MediaTek Research, Senior Technology Manager | Mar. 2024 - Present @Taipei, Taiwan
📌 MediaTek Research, Senior Machine Learning Researcher | Mar. 2023 - Mar. 2024 @Taipei, Taiwan
📌 E.Sun Commercial Bank, Lead Machine Learning Researcher | Feb. 2021 - Mar. 2023 @Taipei, Taiwan
📌 E.Sun Commercial Bank, Machine Learning Engineer | Mar. 2020 - Feb. 2021 @Taipei, Taiwan
📌 Bridgewell, Machine Learning Engineer | Mar. 2018 - Mar. 2020 @Taipei, Taiwan
📌 Ur Warm Pet Cafe, Resident Singer | Mar. 2017 - Mar. 2018 @Hsinchu, Taiwan
📌 TSMC, Photo Lithography Software R&D Engineer | Nov. 2015 - Mar. 2018 @Hsinchu, Taiwan

PUBLICATIONS
📌 Chan-Jan Hsu*, Yi-Chang Chen*, Feng-Ting Liao, Pei-Chen Ho, Yu-Hsiang Wang, Po-Chun Hsu, Da-shan Shiu, “Let's Fuse Step by Step: A Generative Fusion Decoding Algorithm with LLMs for Robust and Instruction-Aware ASR and OCR,” in Findings of ACL, 2025. [Paper][Code]
📌 Liang-Hsuan Tseng*, Yi-Chang Chen*, Kuan-Yi Lee, Da-Shan Shiu, Hung-yi Lee, “TASTE: Text-Aligned Speech Tokenization and Embedding for Spoken Language Modeling,” in Arxiv, 2025. [Paper][Overview]
📌 Feng-Ting Liao*, Yung-Chieh Chan*, Yi-Chang Chen, Chan-Jan Hsu, Da-shan Shiu, “Zero-shot Domain-sensitive Speech Recognition with Prompt-conditioning Fine-tuning,” in ASRU, 2023. [Paper][Code]
📌 Yi-Chang Chen, Yu-Chuan Chang, Yen-Cheng Chang, and Yi-Ren Yeh, “g2pW: A Conditional Weighted Softmax BERT for Polyphone Disambiguation in Mandarin,” in INTERSPEECH, 2022. [Paper][Code]
📌 Yen-Cheng Chang, Yi-Chang Chen, Yu-Chuan Chang, and Yi-Ren Yeh, “SMILE: Sequence-to-Sequence Domain Adaptation with Minimizing Latent Entropy for Text Image Recognition,” in ICIP, 2022. [Paper]
📌 Yi-Chang Chen, Yu-Chuan Chang, Yen-Cheng Chang, and Yi-Ren Yeh, “Traditional Chinese Synthetic Datasets Verified with Labeled Data for Scene Text Recognition,” in DLVDR Workshop on ICPR, 2022. [Paper][Code]
📌 Yi-Chang Chen, Chun-Yen Cheng, Chien-An Chen, Ming-Chieh Sung, and Yi-Ren Yeh, “Integrated Semantic and Phonetic Post-correction for Chinese Speech Recognition,” in ROCLING, 2021. [Paper][Code]