Françoise Beaufays

I'm a Research Scientist at Google. I hold a PhD in EE and a PhD minor in Italian Literature from Stanford University. Most recently, I contributed extensively to optimizing the performance of Google Search by Voice. In general, I am fascinated with the opportunity that large amounts of training data may offer in statistical modeling fields such as speech recognition.

Google Publications

Previous Publications

  •  

    Learning Linguistically Valid Pronunciations from Acoustic Data

    Francoise Beaufays, Ananth Sankar, Shaun Williams, Mitchel Weintraub

    Proc. Eurospeech (2003)

  •  

    Learning Name Pronunciations in Automatic Speech Recognition Systems

    Francoise Beaufays, Ananth Sankar, Shaun Williams, Mitchel Weintraub

    Proc. ICTAI (2003)

  •  

    Speech Recognition Technology

    Francoise Beaufays, Herve Bourlard, Horacio Franco

    The Handbook of Brain Theory and Neural Networks, MIT Press (2003)

  •  

    Using Speech/Non-Speech Detection to Bias Recognition Search on Noisy Data

    Francoise Beaufays, Daniel Boies, Mitchel Weintraub, Qifeng Zhu

    Proc. ICASSP (2003)

  •  

    Porting Channel Robustness Across Languages

    Francoise Beaufays, Daniel Boies, Mitchel Weintraub

    Proc. ICSLP (2002)

  •  

    Speech Recognition Technology

    Francoise Beaufays, Herve Bourlard, Horacio Franco, Nelson Morgan

    Handbook of Brain Theory and Neural Networks, MIT Press (2002)

  •  

    Discriminative Mixture Weight Estimation for Large Gaussian Mixture Models

    Francoise Beaufays, Mitchel Weintraub, Yochai Konig

    Proc. ICASSP (1999)

  •  

    Robust Text-Independent Speaker Identification over Telephone Channels

    Hema Murthy, Francoise Beaufays, Larry Heck

    IEEE Trans. on Speech and Audio Processing, vol. 7 (1999)

  •  

    Robustness of Noisy Speech Features Using Neural Networks

    Mitchel Weintraub, Francoise Beaufays

    Proc. Workshop on Robust Methods for Speech Recognition in Adverse Conditions (1999)

  •  

    DYNAMO: An Algorithm for Dynamic Acoustic Modeling

    Francoise Beaufays, Mitchel Weintraub, Yochai Konig

    Proc. 1998 DARPA Broadcast News Transcription and Understanding Workshop

  •  

    Diagrammatic Methods for Deriving and Relating Temporal Neural Network Algorithms

    Eric Wan

    Lecture notes from the Caianiello Summer School on Adaptive Processing of Sequences, Springer-Verlag (1998)

  •  

    Feature Extraction for Speaker Identification

    Hema Murthy, Francoise Beaufays, Larry Heck, Mitchel Weintraub

    Proc. Signal Processing, Communications and Networking (1997)

  •  

    Neural - Network Based Measures of Confidence for Word Recognition

    Mitch Weintraub, Francoise Beaufays, Ze'ev Rivlin, Yochai Konig, Andreas Stolcke

    Proc. ICASSP (1997)

  •  

    Transformation for Robust Speaker Recognition from Telephone Data

    Francoise Beaufays, Mitchel Weintraub

    Proc. ICASSP (1997)

  •  

    Diagrammatic Derivation of Gradient Algorithms for Neural Networks

    Eric Wan, Francoise Beaufays

    Neural Computation (1996)

  •  

    Feature Extraction and Model Training for Robust Speech Recognition

    Ananth Sankar, Andreas Stolcke, Tom Chung, Leo Neumeyer, Mitchel Weintraub, Horacio Franco, Francoise Beaufays

    Proc. ARPA Speech Recognition Workshop (1996)

  •  

    On the Advantages of the LMS Spectrum Analyzer over Non-Adaptive Implementations of the Sliding-DFT

    Francoise Beaufays, Bernard Widrow

    IEEE Trans. on Circuits and Systems (1995)

  •  

    Orthogonalizing Adaptive Algorithms: RLS, DFT/LMS, and DCT/LMS

    Francoise Beaufays

    Adaptive Inverse Control, B. Widrow and E. Walach (1995)

  •  

    Training Data Clustering for Improved Speech Recognition

    Ananth Sankar, Francoise Beaufays, Vassilios Digalakis

    Proc. Eurospeech (1995)

  •  

    Transform Domain Adaptive Filters: An Analytical Approach

    Francoise Beaufays

    IEEE Trans. on Signal Proc. (1995)

  •  

    Two-Layer Linear Structures for Fast Adaptive Filtering

    Francoise Beaufays

    Ph.D. Thesis, Information Systems Laboratory, Department of Electrical Engineering, Stanford University (1995)

  •  

    A Simple Approach to Derive Gradient Algorithms for Arbitrary Neural Network Structures

    Eric Wan, Francoise Beaufays

    Proc. WCNN (1994)

  •  

    A Unified Approach to Derive Gradient Algorithms for Arbitrary Neural Network Structures

    Francoise Beaufays, Eric Wan

    Proc. ICANN (1994)

  •  

    An Efficient First-Order Stochastic Algorithm for Lattice Filters

    Francoise Beaufays, Eric Wan

    Proc. ICANN (1994)

  •  

    Neural Networks to Load-Frequency Control in Power Systems

    Francoise Beaufays, Yousef Abdel-Magid, Bernard Widrow

    Neural Networks, vol. 7 (1994), pp. 183-194

  •  

    Relating Real-Time Backpropagation and Backpropagation Through-Time: An Application of Flow Graph Interreciprocity

    Francoise Beaufays, Eric Wan

    Neural Computation (1994)

  •  

    Simple Approach to Derive Gradient Algorithms for Arbitrary Neural Network Structures

    Eric Wan, Francoise Beaufays

    Proc. WCNN (1994)

  •  

    Two-Layer Linear Structures for Fast Adaptive Filtering

    Francoise Beaufays, Bernard Widrowe

    Proc. WCNN (1994)

  •  

    Simple Algorithms for Fast Adaptive Filtering

    Francoise Beaufays, Bernard Widrow

    Proc. of the Fifth Workshop on Neural Networks, WNN93/FNN93 (1993)