Mon 11/25

Seminar @ Cornell Tech: Karen Livescu

Embeddings for Spoken Words

Word embeddings have become a ubiquitous tool in natural language processing. These embeddings represent the meanings of written words. On the other hand, for spoken language it may be more important to represent how a written word *sounds* rather than (or in addition to) what it means. For some applications it can also be helpful to represent variable-length acoustic segments corresponding to words, or other linguistic units, as fixed-dimensional vectors. This talk will present work on both acoustic word embeddings and “acoustically grounded” written word embeddings, including their applications for improved speech recognition and search.

Speaker Bio

Dr. Livescu received her B.A. in Physics from Princeton University in 1996 and spent the following year as a visiting student in Electrical Engineering at the Technion, Israel. She received her M.S. and Ph.D. in Electrical Engineering and Computer Science at the Massachusetts Institute of Technology, in 1999 and 2005. In 2005-2007 she was a post-doctoral lecturer in the Electrical Engineering and Computer Science department at the Massachusetts Institute of Technology.

Dr. Livescu’s research interests are in speech and language processing, recently focusing on speech recognition. She is particularly interested in statistical modeling techniques that can take advantage of both large stores of data and knowledge from linguistics and speech science.

Dr. Livescu also has a personally maintained website which can be found at