The Department
The Human Language Technology (HLT) Department is a key competency centre within the Institute, throughout Singaporeand the region. Guided by a passion for natural, human languages and a keen sense of innovation, the HLT department's mission is to ideate and produce state-of-the-art language technologies that drive the development of ground-breaking yet practical applications and services. With its valuable team members, the department will chart new research directions and spearhead technological advancements that facilitate and even transform human communication through new language technologies.     

The HLT Department is led by a group of established scientists and engineers, some of whom are synonymous with their respective areas of specialization.  The fields of expertise include automatic speech recognition, speaker and language recognition, statistical natural language processing, machine translation, and information retrieval technologies.  The many years of innovative HLT activities and the varied backgrounds of the HLT research team contribute to the department's strength and its areas of focus: Asian languages and multilingual computing.

The Department's flagship technologies include text information agent, machine translation for South-East Asian Languages, and the Abacus multilingual speech recognition platform - all of which have been deployed in various commercial applications.

Various technologies of the department have duly received international accolades, including consistent, leading (top three) performances in the NIST international evaluations of speaker recognition, language recognition and rich transcription technologies, and the NTCIR information retrieval international benchmarking.

Key Competencies

• Large Vocabulary Continuous Speech Recognition (LVCSR)- Applications:  Automatic Transcription; Spoken Document Retrieval; Voice Surveillance
• Speaker and Language Recognition (SLR) - Applications: Voice Biometrics; Spoken Document Retrieval; Automatic Call Routing; Rich Transcription
• Multilingual Modeling - Applications: Machine Translation; Multilingual Information Management
• Cross-lingual Information Retrieval -Applications: Information Security; Market and SecurityIntelligence; Financial Information Gathering System

Intellectual Capital
Patents

• US Patent: "Spoken Language Identification System and Methods for Training and Operating Same", USA Provisional Patent No. 60/611,022, filed on Sept 17, 2004.
• US Patent: "Method and Apparatus for Voice Annotation and Retrieval of Multimedia Data", (Patent No: US 6,397,181 B1) granted on 28 May 2002.
• US Patent: "Framework: Music Content Representation in Vector Space for Indexing and Retrieval", P200654/US_P, filed on Sept 07, 2006.
• US Patent: "Apparatus and Method for Speech Utterance Verification", filed on 15 Sep 2006.
• US, SG, EU, CN Patent, "A Method for Extraction of Terms from Large-scale of Text Collections", filed in 2003, 2004 and 2005.
• SG Patent, "A Method of Visualizing Clusters of Large Collection of Text Documents", Granted in 2005.
• SG Patent, "Method and System for Personalized Information Management", Granted in 2006.
• SG Patent, "Method and System for Discovering Knowledge from Text Documents", Granted in 2007.

Publications

• Haizhou Li, Bin Ma, and Chin-Hui Lee, "A Vector Space Modeling Approach to Spoken Language Identification", in IEEE Transactions on Audio, Speech and Language Processing, Vol 15, No. 1, 2007.
• Tin Lay Nwe and Haizhou Li, "Exploring Vibrato-Motivated Acoustic Features for Singer Identification", in IEEE Transactions on Audio, Speech and Language Processing, Vol 15, No. 2, 2007.
• Haizhou Li, Khe Chai Sim, Jin-Shea Kuo, and Minghui Dong, "Semantic Transliteration of Personal Names", 45th Annual Meeting of the Association for Computational Linguistics (ACL), Prague, Czech Republic, June 2007.
• Hendra Setiawan, Min-Yen Kan, Haizhou Li, "Ordering Phrases with Function Words", 45th Annual Meeting of the Association for Computational Linguistics (ACL), Prague, Czech Republic, June 2007.
• Min Zhang, Wanxiang Che, Aiti Aw and Chew Lim Tan, "A Grammar-driven Convolution Tree Kernel for Semantic Role Classification", 45th Annual Meeting of the Association for Computational Linguistics (ACL), Prague, Czech Republic, June 2007.
• Jin-Shea Kuo, Haizhou Li and Ying-Kuei Yang, "Learning Transliteration Lexicons from the Web", 44th Annual Meeting of the Association for Computational Linguistics (COLING-ACL), Sydney, Australia, July 2006.
• Aiti Aw, Min Zhang, Juan Xiao, Jian Su, "A Phrase-based Statistical Model for SMS Text Normalization", 44th Annual Meeting of the Association for Computational Linguistics (COLING-ACL), Sydney, Australia, July 2006.
• Haizhou Li and Bin Ma, "A Phonotactic Language Model for Spoken Language Identification", 43rd Annual Meeting of the Association for Computational Linguistics (ACL), Ann Arbor, USA, June 2005.
• Zheng-Yu Niu, Dong-Hong Ji and Tan Chew Lim, "Word Sense Disambiguation Using Label Propagation Based Semi-supervised Learning Method", 43rd Annual Meeting of the Association for Computational Linguistics (ACL), Ann Arbor, USA, June 2005.
• Lingpeng Yang, Dong-Hong Ji, "Document Reranking by Term Distribution and Maximal Marginal Relevance for Chinese Information Retrieval", Information Processing and Management, 43(2): 315-326, 2007.

Significant Projects/Collaboration

• Centre for Strategic Infocomm Technologies
• Ministry of Manpower
• Singapore Police Force
• Media Development Authority of Singapore
• Zentek Technology Singapore Pte Ltd
• WholeTree Technologies Pte Ltd
• National University of Singapore
• Nanyang Technological University
• Harbin Institute of Technology, China
• National Institute of Information and Communications Technology, Japan

Significant Achievements

• 2nd place in National Institute of Standards and Technology (NIST, US) 2007 Rich Transcription Evaluation international benchmarking
• 2nd place in National Institute of Standards and Technology (NIST, US) 2006 Speaker Recognition Evaluation international benchmarking in 4 categories
• Overall 3rd place in National Institute of Standards and Technology (NIST, US) 2005 Language Recognition Evaluation international benchmarking
• Best Chinese Information Retrieval System, NTCIR 4, NTCIR 5 international benchmarking in 2004 and 2005, respectively
• The Enterprise Challenge Awards 2005 - Intelligent Voice Profiling System
• The Enterprise Challenge Awards 2004 - Voice Enabling Tan Tock Seng Hospital
• NSTB Technology Innovation Awards 1996 - Chinese Dictation Kit

Click here for the Human Language Technology Department infosheet.