OpenFst - A General and Efficient Weighted Finite-State Transducer Library
Center for Language & Speech Processing(CLSP), JHU via YouTube
AI, Data Science & Cloud Certificates from Google, IBM & Meta
Earn Your CS Degree, Tuition-Free, 100% Online!
Overview
AI, Data Science & Cloud Certificates from Google, IBM & Meta — 40% Off
One plan covers every Professional Certificate on Coursera. 40% off Coursera Plus Annual.
Unlock All Certificates
Learn about OpenFst, a comprehensive weighted finite-state transducer library, in this technical lecture delivered by Michael Riley from Google at Johns Hopkins University's Center for Language & Speech Processing. Explore the design principles, implementation details, and practical applications of this general-purpose library that provides efficient algorithms for constructing, combining, optimizing, and searching weighted finite-state acceptors and transducers. Discover how OpenFst serves as a foundational tool for speech recognition, natural language processing, machine translation, and other computational linguistics applications. Gain insights into the library's architecture, its support for various semirings, and the optimization techniques that make it suitable for large-scale language processing tasks. Understand the theoretical foundations of weighted finite-state transducers and their role in modern speech and language technologies through practical examples and implementation considerations.
Syllabus
Michael Riley: "OpenFst: a General and Efficient Weighted Finite-State Transducer Library"
Taught by
Center for Language & Speech Processing(CLSP), JHU