Geoffrey Zweig

Geoffrey is an AI Director and supports the Multimodal Video Understanding team. He focuses on machine learning algorithms that understand the full spectrum of audio and visual content signals. These algorithms enable key applications in search, ranking, content creation, monetization, and content moderation. Prior to Facebook, Dr. Zweig led research teams at IBM, Microsoft, and JPMorgan Chase, in the areas of speech recognition, dialog systems, NLP, and time series analysis. He received his Ph.D. from UC Berkeley in AI, has published over 80 conference and journal papers, holds numerous patents, and is an IEEE Fellow.

Geoffrey's Publications

May 03, 2020

RESEARCH

Multi-modal Content Localization in Videos Using Weak Supervision

Identifying the temporal segments in a video that contain content relevant to a category or task is a difficult but interesting problem. This has applications in fine-grained video indexing and retrieval. Part of the difficulty in this problem…

Gourab Kundu, Prahal Arora, Ferdi Adeputra, Polina Kuznetsova, Daniel McKinnon, Michelle Cheung, Larry Anazia, Geoffrey Zweig,

May 03, 2020

May 03, 2020

RESEARCH

NLP

From Senones to Chenones: Tied Context-Dependent Graphemes for Hybrid Speech Recognition

There is an implicit assumption that traditional hybrid approaches for automatic speech recognition (ASR) cannot directly model graphemes and need to rely on phonetic lexicons to get competitive performance, especially on English which has poor…

Duc Le, Xiaohui Zhang, Weiyi Zhang, Christian Fuegen, Geoffrey Zweig, Michael L. Seltzer,

May 03, 2020