2015/12-2021/03, Senior Technical Expert and Team Lead (Security and AI), Leopard Mobile, Taipei, Taiwan

  • 電腦視覺暨網路安全:應用圖像識別於欺詐廣告及區塊鏈惡意程式分析,廣告彈窗及惡意騷擾電話識別等推薦系統,並整合於Clean Master及Security Master等 Apps(2014/07-2021/03)。

    • R2-D2:搜集200多萬個Android之classes.dex,轉譯為彩色圖片,準確率可達96%。

    • C-3PO:依據使用者行為之Android彈窗推薦系統,提升用戶7日留存率。

    • TWMAN (抬丸郎):基於WINDOWS之開源自動化惡意程式行為分析(2008/11-2011/09)。

  • 語音訊號處理:提升服務型機器人的語音互動體驗 (2019/10-2021/03)。

    • 聲紋識別:整合SincNet模型及Kaldi,提升家用型服務機器人之聲紋準確度從0.88至0.94。

    • 語音去噪:整合DEMUCS為TCN,模型量化73.4至9MB,CPU推論由0.2秒加速至0.08秒。

    • 語音分割:整合sudo-rm自監督模型,模型量化11至3MB,CPU推論由2.4秒加速至0.27秒。

  • 自然語言處理:提升服務型機器人對自然語言的語義理解 (2017/10-2019/10)。

    • 機器閱讀理解:解決服務型機器人需要關鍵字及正規表示式問題 (EM: 0.828,F1: 0.899)。

    • 文本糾錯:基於ConvSeq2Seq及bert,整合NER,提升識別效果(EM: 0.7624,F1: 0.948)。

    • 命名實體識別:整合拼音及分類機制,並重新標註相關數據集(F1: 0.84)。

    • 文本相似度:基於TextCNN優化及提升語音服務問答準確度。

    • 文本分類:基於bert取代騰訊詞向量,優化及提升商場品牌識別分類準確度。

  • Reading and implement the SOTA algorithm from papers to make definition for the real requirements.

  • Plan and evaluate the accuracy and response time by comparison with other products, and iteration update.

  • On the other hand, I have landing the SOTA algorithm of Natural Language Processing to improve the user experience of blockchain ranking analysis, service robots and intelligent voice assistants.

    • Machine Reading Comprehension: fine-tune with the new translation (from English) and DRCD datasets, the results are EM: 0.8289 and F1-Score: 0.8996.

    • Text Correction: integrate with NER, Pinyin and ConvS2S and then improve the response time from 3 sec to 0.03 sec, the results are precision 0.9158 to 0.9415, F1 0.9266 to 0.9483, EM 0.6425 to 0.7624.

    • Name entity recognition: integrated with Pinyin and huggingface. The results are F1-Score: 0.84.

    • Text Similarity: replace TextCNN with Albert, and then improve performance.

    • Text Classification: replace Tencent Embedding to BERT, BiLSTM and Attention.

  • With respect to speech separation, enhancement, speaker recognition and speech recognition, which improve the quality of voice, and integrate with name entity recognition, classification, similarity, correction and machine reading comprehension, and make the robots understand human meaning.

    • Speaker recognition: integrate sincnet with PLDA of kaldi, re-design datasets and model quantization, including the results are hit rate from 0.88 to 0.94.

    • Speech enhancement: replace the LSTM of DEMUCS@FB Separator with TCN, model quantization, the size from 73.4 MB to 9 MB and RTF from 0.2 to 0.08.

    • Speech separation: re-design datasets, improve the performance with Self-Supervised, model quantization, the size from 11 MB to 3 MB, SISDR from 12.28 to 10.99 and rtf from 2.4 to 0.27

    • Speech recognition: re-design datasets, improve the performance based on NVIDIA nemo.

  • Develop cyber threat analysis open source toolkits which are integrated with computer vision and based on vulnerability and malicious attack. Including malware analysis, blockchain analysis, pop-ups recommendations, phone scams, and deceptive advertising to improve the performance of some products (Clean Master and Security Master).

    • R2-D2: convert classes.dex to RGB and store it as a color image with the fixed size.

    • C-3PO: collaborative filtering based on hybrid user behavior analysis.

    • TWMAN: open source malware analysis / digital forensics toolkit.

  • I have presented my research at Defcon, RuxCon, OWASP Appsec, and HITCON. At the same time, I have customs training and support experience with national security departments, governments, finance companies and universities.

  • Here are some related news reports with my develop experence.

    • Defending Taiwan, Fighting the Pandemic with Technology (link)

    • NVIDIA Developer: AI Improves the Frequency and Quality of Mobile App Notifications (link)

    • Career: Machine Learning Expert (link)

2014/07-2015/12, Senior Security Engineer,Verint Systems (Taiwan), Taipei, Taiwan

  • Training, Solution and Maintenance (教育訓練、方案規劃產品維護)

  • Malware Analysis and Cyber Security Research (惡意程式分析、研究)

2014/01-2014/07, Senior Security Engineer, Acer Cyber Security, Taipei, Taiwan

  • Training, Solution and Maintenance (教育訓練、方案規劃產品維護)

  • Malware Analysis and Cyber Security Research (惡意程式分析、研究)

2008/11-2011/09, Assistant Researcher, National Center for High-Performance Computing, Tainan, Taiwan

  • Training, Solution and Maintenance (教育訓練、方案規劃產品維護)

  • Malware Analysis and Cyber Security Research (惡意程式分析、研究)

學士 (B.S.) 2002/09-2006/06

立德大學 資訊工程系

Leader University
Computer Science and Information Engineering

2013/09~2013/09:2012 台法(NSC-INRIA)聯合團隊人員交流計畫
Visiting Scholar, TAO, an Inria Saclay-IDF team in Lri, Univ. Paris-Sud, Dr. Olivier Teytaud

2011/10 ~ 2012/05:2010 台灣與英國頂尖大學前期合作研究計畫 (週遊列校_No.10-英國艾賽克斯大學)
Visiting Scholar, University of Essex, School of Computer Science and Electronic Engineering, Prof. Hani Hagras

Funding support "Type-2 Fuzzy Ontology Model and Its Applications" under the framework of the 2010 Initiative Research Cooperation among Top Universities between UK and Taiwan from Taiwan National Science Council (NSC)

National University of Tainan, Dept. of Computer Science and Information Engineering,

Prof. Chang-Shing Lee, Ontology Applications & Software Engineering Laboratory ( OASE Lab.)


  1. 2013/1018:拜訪駐英國台北代表處科技組

  2. 2013/10/14:拜訪 Imperial College London

  3. 2013/10/12:拜訪新加坡國立大學計算機學系

  4. 2013/09/09:2013 台法(NSC-INRIA)聯合團隊人員交流計畫 II: (I)、(II)、(III)、(IV)

  5. 2013/06/24:南臺灣國際產學研發教育聯盟交流論壇暨平台啟用儀式:(I)、(II)、(III)、(IV)、(V)

  6. 2013/03/11:2013 Taiwanese-French Research Collaboration on Energy Management

  7. 2013/01/21:2013聽覺障礙教育暨校園性平工作坊

  8. 2012/12/14:國科會102年度專題研究計畫申請經驗分享 暨 產學小聯盟意見溝通座談會

  9. 2012/11/26:臺灣魔圍棋邀請賽@黑嘉嘉、周平強 vs 臺灣魔圍棋

  10. 2012/11/26:國立臺南大學 114週年校慶活動-師生研發成果發表會暨年終記者會

  11. 2012/11/19:國立臺南大學姐妹校英國艾塞克斯大學提供學術交流獎學金

  12. 2012/11/17:Human vs. Computer Go Competition @ TAAI 2012

  13. 2012/11/01:國立臺南大學與臺南市立歸仁國民中學專業發展合作簽約儀式

  14. 2012/11/01:國立臺南大學與臺鹽實業股份有限公司產學合作簽約儀式

  15. 2012/08/14:國立臺南大學專業發展合作單位 聯席會議

  16. 2012/07/06:2012南大數位營Google台灣演講

  17. 2012/06/12:Human vs. Computer Go Competition @ IEEE WCCI 2012

  18. 2012/06/08:Acer eDC 資安維運處 處長 至南大數位演講

  19. 2012/05/30:南科產學協會 - 半導體推廣委員會 - 101年度半導體產學論壇

  20. 2012/05/27:雲端奕趣 - 臺灣魔圍棋邀請賽

  21. 2012/05/25:Google台灣 副總 至南大數位系演講

  22. 2012/05/16:2012 French-Taiwanese Workshop on Energy Management

  23. 2012/04/26:南臺灣國立大學研發策略聯盟拜訪駐英國台北代表處

  24. 2012/04/25:南臺灣國立大學研發策略聯盟拜訪 Imperial College London

  25. 2012/04/23:2012 Research Collaboration Workshop for Top UK-TW Universities

  26. 2012/03/01:南臺灣國立大學研發策略聯盟 參訪 Google臺灣暨Google+ Hangouts 平台啟用儀式

  27. 2012/02/03:拜訪 Google UK

  28. 2012/01/12:拜訪 Google 臺灣

  29. 2011/06/27:FUZZ-IEEE 2011_全球電腦圍棋爭霸賽@TCGA

  30. 2011/06/26:FUZZ-IEEE 2011_歐洲+美國+日本-南臺灣學術論壇

  31. 2011/04/14:IEEE SSCI 2011-Computer Go Competition

  32. 2010/11/23:台法國合計畫-整合知識本體於蒙地卡羅樹搜尋與其應用: 2010 臺灣魔圍棋圍棋棋士邀請賽

  33. 2010/07/20:Human vs. Computer Go Competition at WCCI 2010

  34. 2010/04/02:國立臺南大學與國家高速網路與計算中心合作協議備忘錄簽署儀式

  35. 2010/03/21:臺灣魔圍棋圍棋棋士邀請賽

  36. 2009/10/30:TAAI2009-第十四屆人工智慧與應用研討會-電腦圍棋錦標賽-朝陽科技大學

  37. 2009/10/26:全球電腦圍棋賽成果暨臺灣魔圍棋(MoGo-TW)團隊成立記者會

2001 ~ 2002:嘉南藥理科技大學 進修部 工業安全與衛生學系 (週遊列校_No.06-嘉南藥理科技大學)
Chia Nan University, The School of Continuing Education, Department Pharmacy & Science

1999 ~ 2000:力行升大學 保證班 (週遊列校_No.04-力行重考班)

1999 ~ 1999:高雄市私立復華中學 (週遊列校_No.03-復華中學) Kaohsiung Private Fu-Hwa High School

1998 ~ 1999:國立岡山高級中學 (週遊列校_No.02-岡山高中) National Kangshan Senior High School

1996 ~ 1998:高雄市立左營高級中學 (週遊列校_No.01-左營高中) Kaohsiung Municipal Tsoying Senior High School