Acknowledgements

Phonometrica uses (parts of) the following open source libraries, sometimes with modifications:

• Boost (Boost license), see boost.org

• CppAD, by Brad Bell and contributors (EPL-2.0 / GPL-2+), see coin-or.github.io/CppAD

• Eigen, by Benoît Jacob, Gaël Guennebaud and contributors (MPL-2), see eigen.tuxfamily.org

• FIR filter class, by Mike Perkins (BSD-3-Clause), see cardinalpeak.com

• LBFGS++, by Yixuan Qiu, based on work by Naoaki Okazaki (MIT), see lbfgspp.statr.me

• Lucide icons, by Lucide icons and contributors (ISC License), see lucide.dev

• PCRE2, by Philip Hazel (BSD), see github.com/PCRE2Project/pcre2

• pocketfft, by Martin Reinecke (BSD-3-Clause), see gitlab.mpcdf.mpg.de/mtr/pocketfft

• pugixml, by Arseny Kapoulkine (MIT), see pugixml.org

• QScintilla, by Riverbank Computing (GPL-3), see riverbankcomputing.com

• Qt 6, by The Qt Company (LGPL-3 / GPL-2+), see www.qt.io

• r8brain-free-src, by Aleksey Vaneev (MIT), see github.com/avaneev/r8brain-free-src

• REAPER, by David Talkin at Google Inc. (Apache-2.0), see github.com/google/reaper

• RTAudio, by Gary P. Scavone (MIT), see www.music.mcgill.ca/~gary/rtaudio

• sigslot, by Pierre-Antoine Lacaze (MIT) github.com/palacaze/sigslot

• Snack, by Jonas Beskow and Kåre Sjölander (BSD), see www.speech.kth.se/snack/

• sol2, by Rapptz, ThePhD, and contributors, see github.com/ThePhd/sol2

• SPTK, by Keiichi Tokuda, Keiichiro Oura, Takenori Yoshimura, Takato Fujimoto and contributors (Apache License 2.0), see github.com/sp-nitech/SPTK

• SWIPE, by Kyle Gorman (MIT), see github.com/kylebgorman/swipe

• UTF8-CPP, by Nemanja Trifunovic (MIT), see github.com/nemtrif/utfcpp

• utf8proc, by JuliaStrings and the Public Software Group (MIT), see juliastrings.github.io/utf8proc

• zip, by Kuba Podgórski, based on miniz, by Rich Geldreich (public domain), see github.com/kuba–/zip

Phonometrica uses GitHub to host its source code. The source code is available here.

The implementation of Phonometrica’s scripting engine was partly inspired by Robert Nystrom’s excellent book Crafting Interpreters.

Portions of the statistical estimation logic — including negative binomial regression, mixed-effects models, generalized additive models, DHARMa-style residual diagnostics, estimated marginal means, approximate Bayesian inference (INLA-style), WAIC, and PSIS-LOO — were developed with the assistance of Claude Opus 4.6 (Anthropic), based on published statistical literature and reference R implementations. All AI-assisted logic was manually audited, refactored, and validated against reference R packages (glmmTMB, mgcv, emmeans, MuMIn, DHARMa, loo, brms) to ensure mathematical accuracy and implementation integrity.

We are also grateful to JetBrains for providing us with a non-commercial license of their C++ editor.

The development of coding protocols was originally developed as part of the following research project: A corpus-based longitudinal study of the interphonological features of Japanese learners of French. PI: Sylvain DETEY (Waseda University). This project was supported by the Japanese Society for the Promotion of Science (JSPS), Grant-in-Aid for Scientific Research (B) n°23320121 (2011-2014).