Learning Resources

These resources are here to complement and supplement the learning you do in class and through the textbook. Please make liberal use of them! I will update this page with more resources as the semester progresses.

Datasets

Directories

Pitt LibGuides for linguistics data
Linguistic Data Consortium (LDC)
Linguistic Linked Open Data
General-purpose data repositories:
- Open Science Foundation (OSF)
- Kaggle
Increasingly, authors are publishing their data alongside peer-reviewed articles, so that can be a good place to look.

Text corpora

Natural Language Toolkit (NLTK) corpora index (link, GitHub repo)
Pitt English Language Institute Corpus (PELIC)
EnronSent corpus

Speech corpora

Derived data

Datasets from SPeech Across Dialects of English (SPADE) project
World Atlas of Language Structures (WALS) Online
Datasets from published research. Some examples:
- Egger et al. 2020. Improving the robustness of infant lexical processing speed measures. Behavior Research Methods (article, data)
- Koenecke et al. 2020. Racial disparities in automated speech recognition. PNAS (article, data)
- Sonderegger et al. 2017. The medium-term dynamics of accents on reality television. Language (article, data)
- Villarreal et al. 2021. Gender separation and the Speech Community: Rhoticity in early 20th century Southland New Zealand English. Language Variation and Change (article, data)
Datasets from textbook Statistics for linguists: An introduction using R (Bodo Winter, 2019)

Topics

Data visualization

R for data science (class version) chapter 3 on ggplot2 basics
R for data science (class version) chapter 7 on exploratory data analysis via visualization
Intro to ggplot2 by Joey Stanley
Making vowel plots in R by Joey Stanley (Note: The read.csv() line from this tutorial won’t work on your computer. To read the data, click here to download, then open the file with Rstudio. The data will load as an R object named darla)
ggplot2 cheatsheet

Data wrangling

R for data science (class version), chapters 9–16
Hadley Wickham’s J Stat Soft paper Tidy data: a more theoretical approach to the problem of data tidiness
Project-oriented workflow blog post by Jenny Bryan

Regular expressions

Regular expressions tutorials
Regex crossword
stringr cheatsheet, with regex on page 2

Open Access, data publishing

Justin Kitzes. 2018. The Basic Reproducible Workflow Template. In Justin Kitzes, Daniel Turek, Fatma Deniz (Eds.) The Practice of Reproducible Research. [link]
Copyright and Intellectual Property Toolkit by Lauren Collister [link]
Data Sharing for Linguists, guest lecture by Pitt’s Lauren Collister [slides]
D-Scholarship @ Pitt: Institutional Repository at the University of Pittsburgh [link]

Web scraping

rvest package
SelectorGadget: gadget site, how to use in conjunction with rvest article
Fun way to learn/practice CSS selectors: CSS dinner

Text processing

Text Mining with R textbook covers “tidy text” data structure, sentiment analysis, frequency (TF-IDF), ngrams and correlations, topic modeling, and 3 case studies
- Companion package: tidytext. Uses “tidy data” principles and plays nice with tidyverse packages.
quanteda package for “quantitative analysis of textual data”. Fairly widely used, and also good for “tidy data” principles
spacyr wraps the popular and powerful Python spaCy library
- More powerful than quanteda, but relies on installing a bunch of stuff (incl. Python!)
- Parsing, lemmatizing, dependencies, entity extraction
- Interfaces nicely with tidytext
Older-school R packages for doing text analysis: tm and openNLP. These don’t use “tidy data” principles, but might have more community support resources (e.g., StackOverflow)

Machine learning

James et al 2021 textbook An introduction to statistical learning: with Applications in R (2nd ed.). Website includes full text (!), R Markdown files for each chapter, and other resources
R caret package manual

Software tools

Text editors

Atom recommended for all systems.
Notepad++ (Windows only)
- User defined languages for custom syntax highlighting
Sublime Text (any OS)

Git & GitHub

Git download & installation
LSA 2019 Reproducible Research Workshop tutorials by Pitt’s Na-Rae Han (forked and lightly adapted by Dan): Part 1 Intro to Git, Part 2 Linking Git with GitHub
Introduction to Git course on DataCamp
YouTube video How to get started with Git and GitHub
Git - the simple guide
Git cheatsheet

Markdown

R

Numerous R courses on DataCamp (free to use with your Pitt/CMU login)

Tidyverse

Tidyverse cheatsheets: 1–2-page reference cards for various tidyverse packages (even more here)
DataCamp Introduction to the tidyverse course

R Markdown

R Markdown cheatsheet
DataCamp Reporting with R Markdown course
R Markdown: The definitive guide: Book-length treatment of Rmd with a focus on different output formats
R Markdown cookbook: Another book by the creator of Rmd, this one focusing on shorter examples

Speech data

Praat: Speech analysis and synthesis software
Will Styler’s popular Using Praat for Linguistic Research handbook
Praat scripting tutorials:
- Original
- Phonetics on speed
- Daniel Riggs’s tutorial—has files for download!
- Section 11 (starting on p. 55) of Will Styler’s Using Praat for Linguistic Research handbook (files can be downloaded here)
Notepad++ syntax highlighting file for Praat scripting language (somewhat out of date, but still very helpful)
Praat script collections:
Montreal Forced Aligner: Automatic segmental alignment (GitHub)
Kaldi ASR toolkit
LaBB-CAT corpus analysis tool
- Demo corpus with worksheets for self-directed exploration of LaBB-CAT’s functionality
- R package and Python library for interacting with corpora hosted on LaBB-CAT