李琳山 Lin-shan Lee

國立台灣大學電機工程學系 教授
Professor, Department of Electrical Engineering, National Taiwan University

主要研究領域:

數位語音處理

Major Research Areas:

Computer Processing of Speech Signals

研究領域摘要:

  1. 聲學辨識：聲調模型與辨識、單位選擇與聲學模型、關鍵詞擷取、雜訊處理及通道效應、語者調適、語者確認等。
  2. 語言處理：語言模型、中文詞群研究、對不同主題領域之分析、模型調適、語意分析及語言理解、文件分類等
  3. 聲學辨識與語言處理之整合：發音模型及詞典調適、自發性語音處理、音節為基礎之搜尋、韻律模型、語音理解、對話模型及分析、語音為基礎之資訊檢索等。
  4. 學術性雛型系統：聽寫系統、文字轉語音系統、對話系統、語音為基礎之資訊檢索系統等。
  5. 網路環境之研究方向：主從式架構、網路之語音介面、隨網路資源調適之詞典和語言模型、網路語料處理、無線環境下之分散式語音處理等。

Research Summary:

  1. Acoustic Recognition: tone modeling and recognition, unit selection and acoustic modeling, keyword spotting, handling noise and channel effect, speaker adaptation, speaker verification, etc.
  2. Linguistic Processing: language modeling, Chinese word classification, analysis with respect to different subject domains, model adaptation, semantic analysis and language understanding, PAT-tree-based approaches, segment pattern and key phrase extraction, document classification, etc.
  3. Integrated Acoustic/ Linguistic Processing: pronunciation modeling and lexicon adaptation, spontaneous and telephone speech processing, syllable-based one-pass search, prosody modeling, speech understanding, dialogue modeling and analysis, speech-based retrieval models and features, etc.
  4. Academic Prototype Systems: dictation systems, text-to-speech systems, dialogue systems, speech-based information retrieval systems, etc.
  5. Directions towards Network Environment: client-server architecture, speech human-network interface, network resource adapted lexicon and language model, network corpora processing, distributed speech processing under wireless environment, etc.
Photo of Lin-shan Lee

代表性著作 Selected Publication

  1. Chia-Yu Wan, Lin-Shan Lee, “Histogram-Based Quantization for Robust and/or Distributed Speech Recognition,” IEEE Transactions on Audio, Speech and Language Processing, Volume 16, Issue 4, pp.859–873, May 2008
  2. Ming-Yi Tsai, Fu-Chiang Chou, Lin-shan Lee, “Pronunciation Modeling with Reduced Confusion for Mandarin Chinese Using A Three-stage Framework,” IEEE Transactions on Audio, Speech and Language Processing, Vol.15, No.2, pp.661-675, Feb. 2007
  3. Gwo-Hwa Ju, Lin-shan Lee, “A Perceptually Constrained GSVD-based Approach for Enhancing Speech Corrupted by Colored Noise,” IEEE Transactions on Audio, Speech and Language Processing, Vol. 15, No.1, pp.119-134, Jan. 2007
  4. Lin-shan Lee, Sheng-yi Kong, Yi-cheng Pan, Yi-sheng Fu, Yu-tsun Huang, “Multi-layered Summarization of Spoken Document Archives by Information Extraction and Semantic Structuring,” Interspeech Conference, International Speech Communication Association (ISCA), 1539-1542, Pittsburgh, USA, Sept. 2006
  5. Jeih-weih Hung and Lin-shan Lee, “Optimization of Temporal Filters for Constructing Robust Features in Speech Recognition,” IEEE Transactions on Speech and Audio Processing, Vol.14, No.3, pp.808-832, May 2006
  6. Lin-shan Lee and Berlin Chen, “Spoken Document Understanding and Organization,” IEEE Signal Processing Magazine, Vol. 22, No.5, pp.42-60, Sept. 2005
  7. Yu Tsao, Shang-ming Lee and Lin-shan Lee, “Segmental Eigenvoice with Delicate Eigenspace for Improved Speaker Adaptation,” IEEE Transactions on Speech and Audio Processing, Vol.13, No.3, pp.399-411, May 2005
  8. Berlin Chen, Hsin-Min Wang and Lin-shan Lee, “Discriminating Capabilities of Syllable-based Features and Approaches of Utilizing Them for Voice Retrieval of Speech Information in Mandarin Chinese,” IEEE Transactions on Speech and Audio Processing, Vol.10, No.5, pp.303-314, Jul. 2002
  9. Lin-shan Lee, Yumin Lee, “Voice Access of Global Information for Broadband Wireless: Technologies of Today and Challenges of Tomorrow, (invited paper),” Proceedings of the IEEE, pp. 41-57, Jan. 2001