Explore the latest research from Meta

All Publications

December 17, 2021
Xiaohui Zhang, Vimal Manohar, David Zhang, Frank Zhang, Yangyang Shi, Nayan Singhal, Julian Chan, Fuchun Peng, Yatharth Saraf, Michael L. Seltzer

In this paper, by decoupling the concepts of modeling units and label topologies and building proper numerator/denominator graphs accordingly, we establish a generalized framework for hybrid acoustic modeling (AM).

December 14, 2021
Donglai Xiang, Fabián Prada, Timur Bagautdinov, Weipeng Xu, Yuan Dong, He Wen, Jessica Hodgins, Chenglei Wu

We then train a new two-layer codec avatar with separate modeling of the upper clothing and the inner body layer. To learn the interaction between the body dynamics and clothing states, we use a temporal convolution network to predict the clothing latent code based on a sequence of input skeletal poses. We show photorealistic animation output for three different actors, and demonstrate the advantage of our clothed-body avatars over the single-layer avatars used in previous work.

December 13, 2021
Vimal Manohar, Tatiana Likhomanenko, Wei-Ning Hsu, Ronan Collobert, Yatharth Saraf, Geoffrey Zweig

In this paper, we introduce the Kaizen framework that uses a continuously improving teacher to generate pseudo-labels for semi-supervised speech recognition (ASR). The proposed approach uses a teacher model which is updated as the exponential moving average (EMA) of the student model parameters.