Baum-Welch expectation maximization library
Version 0.3.8 is now available for download.
Baum-Welch is now available on
conda-forge. Linux (x86) and Mac OS X users who already have
conda can install the library and all dependencies using the following command:
conda install -c conda-forge baumwelch
.
OpenGrm Baum-Welch is a C++ library (including associated binaries) which allows the user to estimate the parameters of a discrete hidden Markov model (HMM)
using the Baum-Welch algorithm (a special case of the expectation maximization meta-algorithm). It uses
OpenFst library
finite-state transducers (FSTs) and FST archives (FARs) as inputs and outputs.
If you use this toolkit in your research, we would appreciate it if you cited at least one of:
K. Gorman and C. Allauzen. 2024.
A* shortest string decoding for non-idempotent semirings. In
Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics, in press. (Describes the decoding algorithm.)
K. Gorman, C. Kirov, B. Roark, and R. Sproat. 2021.
Structured abbreviation expansion in context. In
Findings of the Association for Computational Linguistics: EMNLP 2021, pages 995-1005. (Describes the pair n-gram formulation.)