Statistical Machine Translation: The Basic, the Novel, and the Speculative

Philipp Koehn

Downloads

pdf presentation slides (0.6 MB)

flv flash video (530 MB, 320 * 256 pixels)

mpeg2 video part 1  (1400 MB, 720 * 576 pixels)

mpeg2 video part 2  (1300 MB, 720 * 576 pixels)

Abstract

Statistical machine translation has matured in recent years into a viable challenge to traditional, more knowledge-driven methods, with a energetic research community and commercial interest. This tutorial serves as a basic introduction to the basic principles of current statistical machine translation methods, present the latest research in the field and point to the challenges and ideas beyond.

Specifically, the tutorial covers:

• Foundations: available data and tools, generatice modeling, EM training, phrase-based models, evaluation.
• Advanced Methods: log-linear models, discriminative training, specialized models, integration with speech
• Outlook: Syntax-based and syntax-aided approaches

Related link: Upcoming Textbook on Statistical Machine Translation