Vincent Vanhoucke

Vincent Vanhoucke is a Principal Scientist at Google. He is a technical lead and manager in Google's deep learning infrastructure team. Prior to that, he lead the speech recognition quality effort for Google Search by Voice. He holds a Ph.D. in Electrical Engineering from Stanford University and a Diplôme d'Ingénieur from the Ecole Centrale Paris.

Google Publications

Previous Publications

  •  

    Autoregressive Product of Multi-frame Predictions Can Improve the Accuracy of Hybrid Models

    Navdeep Jaitly, Vincent Vanhoucke, Geoffrey Hinton

    Proceedings of Interspeech 2014

  •   

    Reading Text in Consumer Digital Photographs

    Vincent Vanhoucke, S. Burak Gokturk

    Proceedings of SPIE DRR XIV (2007)

  •   

    Automatic Training Set Segmentation For Multi-Pass Speech Recognition

    Mark Z. Mao, Vincent Vanhoucke, Brian Strope

    Proceedings of ICASSP 2005

  •   

    Confidence Scoring and Rejection using Multi-Pass Speech Recognition

    Vincent Vanhoucke

    Proceedings of Interspeech 2005

  •   

    Design of Compact Acoustic Models through Clustering of Tied-Covariance Gaussians

    Mark Z. Mao, Vincent Vanhoucke

    Proceedings of ICSLP 2004

  •   

    Mixtures of Inverse Covariances

    Vincent Vanhoucke, Ananth Sankar

    IEEE Transactions on Speech and Audio Processing, vol. 13 (2004), pp. 250-264

  •   

    Interpretability in Multidimensional Classification

    Vincent Vanhoucke, Rosaria Silipo

    Interpretability Issues in Fuzzy Modeling, Springer-Verlag (2003), pp. 193-217

  •   

    Mixtures of Inverse Covariances

    Vincent Vanhoucke, Ananth Sankar

    Proceedings of ICASSP2003, also in Proceedings of NNSP 2003

  •   

    Mixtures of Inverse Covariances: Covariance Modeling for Gaussian Mixtures with Applications to Automatic Speech Recognition

    Vincent Vanhoucke

    Ph.D. Thesis, Stanford University (2003)

  •   

    Variable Length Mixtures of Inverse Covariances

    Vincent Vanhoucke, Ananth Sankar

    Processings of Eurospeech 2003

  •   

    Effects of Prompt Style when Navigating through Structured Data

    Vincent Vanhoucke, W. Lawrence Neeley, Maria Mortati, Michael J. Sloan, Clifford Nass

    Proceedings of INTERACT 2001, Eighth IFIP TC.13 Conference on Human Computer Interaction, pp. 530-536

  •   

    Speaker-Trained Recognition using Allophonic Enrollment Models

    Vincent Vanhoucke, Michael M. Hochberg, Christopher J. Leggetter

    Proceedings of ASRU2001