The IBM 2016 English Conversational Telephone Speech Recognition System

George Saon, Tom Sercu, Steven Rennie +1

English Conversational Telephone Speech Recognition by Humans and Machines

George Saon, Gakuto Kurata, Tom Sercu +9

Sobolev Descent

Youssef Mroueh, Tom Sercu, Anant Raj

The paper introduces Sobolev Descent, a method that simplifies GAN training by transporting particles from a source to a target distribution using gradient flows of a Sobolev GAN c…

Wasserstein Barycenter Model Ensembling

Pierre Dognin, Igor Melnyk, Youssef Mroueh +3

Advances in Very Deep Convolutional Neural Networks for LVCSR

Tom Sercu, Vaibhava Goel

Very Deep Multilingual Convolutional Neural Networks for LVCSR

Tom Sercu, Christian Puhrsch, Brian Kingsbury +1

Multi-Frame Cross-Entropy Training for Convolutional Neural Networks in Speech Recognition

Tom Sercu, Neil Mallinar

PepCVAE: Semi-Supervised Targeted Design of Antimicrobial Peptide Sequences

Payel Das, Kahini Wadhawan, Oscar Chang +6

Adversarial Semantic Alignment for Improved Image Captions

Pierre L. Dognin, Igor Melnyk, Youssef Mroueh +2

Semi-Supervised Learning with IPM-based GANs: an Empirical Study

Tom Sercu, Youssef Mroueh

Sobolev GAN

Youssef Mroueh, Chun-Liang Li, Tom Sercu +2

Fisher GAN

Youssef Mroueh, Tom Sercu

Dense Prediction on Sequences with Time-Dilated Convolutions for Speech Recognition

Tom Sercu, Vaibhava Goel

Big-Little Net: An Efficient Multi-Scale Feature Representation for Visual and Speech Recognition

Chun-Fu Chen, Quanfu Fan, Neil Mallinar +2

The paper introduces Big-Little Net, a CNN architecture that combines multiple branches with different computational costs to efficiently capture multi-scale features, improving sp…

McGan: Mean and Covariance Feature Matching GAN

Youssef Mroueh, Tom Sercu, Vaibhava Goel