Research Interests

• Speech Recognition: acoustic modeling
• Web-scale natural language processing
• Audio-Video navigation and personalization (e.g. for ads)
• Machine Translation

Background

In 1998, Patrick Nguyen received an Engineer Diploma in Telecommunication Systems from EPFL (Swiss Federal Institute for Technology), winning the Hitachi Award. In 1999, he co-founded a company called BeTrust, whose main task was to build a trading platform for RealtimeForex. In 2002, he received a Doctorate in Technical Science from the same university. He served as a Senior Engineer in Panasonic (Panasonic Speech Technology Laboratory or PSTL, in Santa Barbara, CA), joining the company in 2000 and finally parting ways in 2004. It was during that time that he initiated and lead the large vocabulary speech recognition effort. He is most proud of being an early contributor to Eigenvoices with R. Kuhn, achieving the best results on Aurora4 with L. Rigazio in 2003, building the first large corpus (1000h+) system in a NIST evaluation system in 2003, and obtaining the best speaker diarization results in the NIST RT02 and RT03 evaluations, with Y. Moh. He was awarded about 12 patents including co-authorship.
In June 2004, he joined Microsoft Research where he surrendered his lust and appetite for closed-form solutions.

He is occasionally seen (or even scobleized, see 6) to present useful work.

He's been working on review summarization. Call our VoiceRate system: 1-877-456-DATA.

He also released a Scalable Language Modeling Toolkit, Microsoft Research Language Modeling (MSRLM, download here). The toolkit implements an efficient method to build large language models, from billions of words and upwards. We use these language models for first-pass decoding in statistical machine translation.

Publications

• G. Zweig, P. Nguyen, Y.-C. Ju, Y.-Y. Wang, D. Yu, and A. Acero. The Voice-Rate Dialog System for Consumer Ratings. In Interspeech, 2007.
• P. Nguyen, M. Mahajan, and X. He. Training Non-Parametric Features for Statistical Machine Translation. In Second Workshop on Statistical Machine Translation(ACL), 2007.
• G. Zweig, Y. C. Ju, P. Nguyen, D. Yu, Y.-Y. Wang, and A. Acero. Voice-Rate: A Dialog System for Consumer Ratings. In HLT, 2007 (demo track).
• A. Subramanya, Z. Zhang, A. C. Surendran, P. Nguyen, M. Narasimhan, and A. Acero. A Generative-Discriminative Framework using Ensemble Methods for Text-Dependent Speaker Verification. In ICASSP, 2007.
• C. Ma, P. Nguyen, and M. Mahajan. Finding Speaker Identities with a Conditional Maximum Entropy Model. In ICASSP, 2007.
• P. Nguyen, and M. Mahajan. Audio/Video Navigation with A/V X-Ray. In SLT, 2006 (demo track).
• X. He, A. Menezes, C. Quirk, A. Aue, S. Corston-Oliver, JF. Gao, and P. Nguyen. Microsoft Research Treelet Translation System: NIST MT Evaluation 06. In NIST Machine Translation Workshop, 2006.
• P. Nguyen. Panasonic Real-Time Meeting Room STT. In NIST Rich Transcription (Spring), 2004.
• Y. Moh, P. Nguyen, and J.-C. Junqua. Towards Domain Independent Speaker Clustering. In ICASSP, 2003.
• P. Nguyen, L. Rigazio, and J.-C. Junqua. Large Corpus Experiment for Broadcast News Recognition. In Proceedings of Eurospeech, 2003.
• L. Rigazio, P. Nguyen, D. Kryze, and J.-C. Junqua. Large Vocabulary Noise Robustness on Aurora4. In Proceedings of Eurospeech, 2003.
• P. Nguyen and J.-C. Junqua. PSTL's Speaker Diarization system. In DARPA/NIST Rich Transcription Workshop, 2003.
• P. Nguyen and J.-C. Junqua. PSTL's Speech-to-Text system. In DARPA/NIST Rich Transcription Workshop, 2003.
• P. Nguyen SWAMP: An Isometric Frontend for Speaker Clustering. In DARPA/NIST Rich Transcription Workshop, 2003.
• P. Nguyen, L. Rigazio, C. Wellekens, and J.-C. Junqua. LU Factorization for Feature Transformation. In ICSLP, 2002.
• P. Nguyen, L. Rigazio, J.-C. Junqua, and C. Wellekens. Piecewise Linear Constraints for Model Space Adaptation . In ICASSP, 2002.
• Y. Souilmi, L. Rigazio, P. Nguyen, D. Kryze, and J.-C. Junqua. Blind channel estimation based on speech correlation structure. In ICASSP, 2002.
• P. Nguyen, L. Rigazio, Y. Moh, and J.-C. Junqua. Rich Transcription 2002: Site Report (PSTL). In NIST Rich Transcription Workshop, 2002.
• P. Nguyen, L. Rigazio, C. Wellekens, J.-C. Junqua. Construction of Model-Space Constraints. In ASRU, 2001.
• R. Kuhn, F. Perronnin, P. Nguyen, J.-C. Junqua, and L. Rigazio. Very Fast Adaptation with a Compact Context-Dependent Eigenvoice Model. In ICASSP, 2001.
• F. Perronnin, R. Kuhn, P. Nguyen, and J.-C. Junqua. Maximum-Likelihood Training of a Bipartite Acoustic Model for Speech Recognition. In ICASSP, 2001.
• L. Rigazio, P. Nguyen, D. Kryze, and Jean-C. Junqua. Separating Speaker and Environment Variabilities for Improved Recognition in Non-Stationary Conditions. In Eurospeech, 2001.
• P. Nguyen, L. Rigazio, R. Kuhn, J.-C. Junqua, and C. Wellekens. Self-Adaptation Using Eigenvoices for Large-Vocabulary Continuous Speech Recognition. In ITRW on Adaptation (ISCA Workshop), 2001.
• R. Kuhn, J.-C. Junqua, P. Nguyen, and N. Niedzielski. Rapid Speaker Adaptation in Eigenvoice Space. In IEEE Transactions on Speech and Audio Processing, VOL. 8, NO. 6, November 2000.
• O. Thyes, R. Kuhn, P. Nguyen and J.-C. Junqua. Speaker Identification and Verification using Eigenvoices. In ICSLP, 2000.
• P. Nguyen, L. Rigazio, and J.-C. Junqua. EWAVES: An Efficient Decoding Algorithm for Lexical Tree Based Speech Recognition. In ICSLP, 2000.
• R. Kuhn, P. Nguyen, J.-C. Junqua, R. Boman, N. Niedzielski, S. Fincke, K. Field, and M. Contolini. Fast Speaker Adaptation using A Priori Knowledge. In ICASSP, 1999.
• P. Nguyen, P. Gelin, J.-C. Junqua, and J.-T. Chien. N-Best Based Supervised and Unsupervised Adaptation for Native and Non-Native Speakers in Cars. In ICASSP, 1999.
• P. Nguyen, C. Wellekens, and J.-C. Junqua. Maximum likelihood eigenspace and MLLR for speech recognition in noisy environments. In Eurospeech, 1999.
• R. Kuhn, P. Nguyen, J.-C. Junqua, L. Goldwasser, N. Niedzielski, S. Fincke, K. Field, and M. Contolini. Eigenvoices for Speaker Adaptation. In ICSLP, 1998.
• R. Kuhn, P. Nguyen, J.-C. Junqua, L. Goldwasser, N. Niedzielski, S. Fincke, and K. Field. Eigenfaces and Eigenvoices: Dimensionality Reduction for Specialized Pattern Recognition. In MMSP, 1998.
• P. Nguyen. Fast Speaker Adaptation. Master's thesis, 1998.