Seminar für Sprachwissenschaft

Our group develops and maintains some free and open source R and Python packages. A short description of the different packages and links to the download side follows.


Coding Sound Files for Use with NDL

Maintainer: Denis Arnold
Authors: Denis Arnold, Elnaz Shafaei Bajestan



Make acoustic cues to use with the R packages 'ndl' or 'ndl2' or Python package 'pyndl'. The package implements functions used in the PLoS ONE paper: Denis Arnold, Fabian Tomaschek, Konstantin Sering, Florence Lopez, and R. Harald Baayen (2017). Words from spontaneous conversational speech can be recognized with human-like accuracy by an error-driven learning algorithm that discriminates between meanings straight from smart acoustic features, bypassing the phoneme as recognition unit. PLoS ONE 12(4):e0174623 <doi:10.1371/journal.pone.0174623> More details can be found in the paper and the supplement.



Interpreting Time Series and Autocorrelated Data Using GAMMs

Maintainer: Jacolien van Rij
Authors: Jacolien van Rij, Martijn Wieling, R. Harald Baayen, Hedderik van Rijn



GAMM (Generalized Additive Mixed Modeling; Lin & Zhang, 1999) as implemented in the R package 'mgcv' (Wood, S.N., 2006; 2011) is a nonlinear regression analysis which is particularly useful for time course data such as EEG, pupil dilation, gaze data (eye tracking), and articulography recordings, but also for behavioral data such as reaction times and response data. As time course measures are sensitive to autocorrelation problems, GAMMs implements methods to reduce the autocorrelation problems. This package includes functions for the evaluation of GAMM models (e.g., model comparisons, determining regions of significance, inspection of autocorrelational structure in residuals) and interpreting of GAMMs (e.g., visualization of complex interactions, and contrasts).


An efficient implementation of Naive Discriminative Learning in Python

Maintainer: Konstantin Sering, Marc Weitz
Authors: Konstantin Sering, Marc Weitz, David-Elias Künstle, Lennart Schneider



pyndl is an implementation of Naive Discriminative Learning in Python. It was created to analyse huge amounts of text file corpora. Especially, it allows to efficiently apply the Rescorla-Wagner learning rule to these corpora.


Naive Discriminative Learning


Maintainer: Konstantin Sering
Authors: Antti Arppe, Peter Hendrix, Petar Milin, R. Harald Baayen, Konstantin Sering, Cyrus Shaoul



Naive discriminative learning implements learning and classification models based on the Rescorla-Wagner equations and their equilibrium equations.


Introduction to Hierarchical MPT Modeling

Authors:  Daniel W. Heck, Nina R. Arnold, Denis Arnold