---+ <nop> SFst: Stochastic Finite-State Transducer Library --- %ICON{led-red}% *OpenGrm SFst version 1.2.1 is now available for [[SFstDownload][download]].* --- _SFst_ is a library for normalizing, sampling, combining, and approximating _stochastic_ (or _probabilistic_) finite-state transducers. These are weighted finite-state transducers, represented in [[http://www.openfst.org][OpenFst library]] format, that have two properties: 1 a [[SFstGlossary#CanonicalDef][canonical]] format that may include [[FST.FstGlossary#EpsilonDef][epsilon]] and [[SFstGlossary#FailureTransitionDef][failure transitions]] under specified constraints 1 a [[SFstGlossary#NormalizedDef][normalized]] weight distribution that assigns a (negative log) probability to each path leaving a state. An n-gram model produced by the [[NGramLibrary][OpenGrm NGram Library]] is a stochastic FST<sup>1</sup> but many other topologies are possible. * [[SFstBackground][Background Material]] * [[SFstQuickTour][Quick Tour]] * [[SFstAvailableOperations][Available Operations]] * [[SFstGlossary][Glossary]] * [[SFstDownload][Download]] * [[http://www.opengrm.org/doxygen/sfst/html/][Documented Source Code]] --- <sup>1</sup>Provided the failure label (=phi_label=) is specified to match the backoff label, typically 0, of the n-gram model.
This topic: GRM
>
WebHome
>
SFstLibrary
Topic revision: r23 - 2024-01-25 - KyleGorman
Copyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki?
Send feedback