communication: Feature Extraction and Model Estimation for Audio of Human Speech

Provides fast, easy feature extraction of human speech and model estimation with hidden Markov models. Flexible extraction of phonetic features and their derivatives, with necessary preprocessing options like feature standardization. Communication can estimate supervised and unsupervised hidden Markov models with these features, with cross validation and corrections for auto-correlation in features. Methods developed in Knox and Lucas (2021) <doi:10.7910/DVN.8BTOHQ>.

Version: 0.1
Depends: R (≥ 3.5.0)
Imports: Rcpp (≥ 1.0.2), purrr, magrittr, diagram, GGally, grid, useful, ggplot2, reshape2, tuneR, wrassp, gtools, signal, plyr, RColorBrewer, scales, abind, igraph, gtable
LinkingTo: Rcpp, RcppArmadillo (≥ 0.9.700.2.0)
Suggests: knitr, qpdf, rmarkdown, testthat
Published: 2021-02-25
DOI: 10.32614/CRAN.package.communication
Author: Dean Knox [aut], Christopher Lucas [aut, cre], Guilherme Duarte [ctb], Alex Shmuley [ctb], Vineet Bansal [ctb], Vadym Vashchenko [ctb]
Maintainer: Christopher Lucas <christopher.lucas at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Materials: README
CRAN checks: communication results


Reference manual: communication.pdf
Vignettes: extracting-features


Package source: communication_0.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): communication_0.1.tgz, r-oldrel (arm64): communication_0.1.tgz, r-release (x86_64): communication_0.1.tgz, r-oldrel (x86_64): communication_0.1.tgz


Please use the canonical form to link to this page.