Lars Hinrichs's Academic Website

05-04-2026

Automated Vowel Formant Extraction with AI

For linguists working with Claude: I have released formant-extraction-skill, a skillset that takes a monolingual American English interview recording from raw audio to a vowel-formant CSV in a single prompt. The pipeline runs Whisper + ECAPA-TDNN diarization, FAVE/MFA or BAS WebMAUS forced alignment, and FAVE-extract or a headless Praat script — driven by natural-language prompts rather than command-line invocations.

The skillset runs in either Claude Cowork (the desktop app) or Claude Code (the CLI/IDE extension). Two human-in-the-loop gates pause for transcript correction and alignment review before the pipeline continues.

To try it:

Clone github.com/wordsmith189/formant-extraction-skill.
Symlink the setup skill into your Claude skills directory:

ln -s "$(pwd)/formant-extraction-setup" ~/.claude/skills/formant-extraction-setup

In Claude, prompt: Run /formant-extraction-setup for the FAVE/MFA path.
When setup completes, prompt: Run /formant-extraction on interview.wav end-to-end.

The README has sample prompts for partial pipelines (e.g. if you already have a TextGrid transcript and only need alignment + extraction), batch-processing guidance, and citation info. v1 covers English only; bilingual support is planned for v2.

01-25-2026

New bachelor degree: B.S. in Behavioral and Social Data Science

Humanities Track Starting in Fall 2026

The College of Liberal Arts (CoLA) at UT Austin is starting a new degree that I played a part in designing: a new humanities track to the nearly-as-new bachelor's degree in Behavioral and Social Data Science. If you are interested, check out info about the program here, and see the list of involved faculty for the psychology and the humanities tracks here.

10-17-2025

E 388D | Text in Code: Digital Methods for Language and Culture

Graduate course, Fall 2026

This course is for students from linguistics, literature, and others who study textual culture. It introduces computational methods for the analysis of textual data, with a focus on scripting in the R programming language. Assuming no prior programming experience, the course emphasizes hands-on learning in corpus building, text mining, statistical analysis, and data visualization. Students will engage with sample studies from synchronic and diachronic linguistics as well as digital humanities, exploring how digital tools can illuminate patterns in language and culture. While instruction centers on R, students will also explore LLM chatbots to support their coding practice. Grading based on readings, short presentations, and an original research project.

Linguistic studies on our reading list will include corpus linguistic approaches to variation and alternation in Present-Day English as well as corpus studies of language change. On the cultural front, we will start by examining distant reading and its recent refinements, then look at emerging trends in DH text analysis using topic modeling, stylometry, sentiment analysis, corpus-based discourse analysis, and machine learning techniques for interpreting narrative and genre.

02-04-2025

Press Coverage for Y'All

In early December, my famous colleague John McWhorter wrote in his NYT column about the pronoun y'all, with muted enthusiasm. Speaking of the lack of a distinct pronoun for the second person plural, John noted that

[some] critics advise us to substitute “you all,” “all,” “folks” or “people.” Some have even suggested “y’all,” a word that reads as much too slangy, regional or what you might even call ethnic to ever gain universal acceptance.

The magazine Texas Monthly felt called upon to rectify this slight on y'all's reputatiton. Initially, shooting somewhat from the hip in this piece that could have used some input (or at least fact-checking) from a linguist.

In the February '25 issue, David Courtney now follows up by dedicating his "Texanist" column to the issue, on the occasion of a reader question that he received. In it, I am quoted as saying that y'all is proof that "Texas English is superior" to Mainstream U.S. English, something I have wanted to say in a media piece for a long time. Read Courtney's piece below, starting with UT's publicity for it.

12-04-2024

Virtual Keynote at COMSATS Lahore, Pakistan

I was invited to give a recorded keynote lecture at this years 2nd International Conference on Modern Trends in Applied Linguistics, hosted by COMSATS University Islamabad, Lahore campus. The title of my presentation is "Cultural Contact and New Dialect Emergence in Large Cities: From Ethnolects to Multiethnolects".

12-02-2023

ORCID ID

I have joined the ORCID initiative. Visit my profile at the URL
orcid.org/0000-0003-3679-1927.

11-29-2023

Sentiment in Shakespeare

In 2020 I was getting accustomed to digital text analysis in R (for much of it, guided by Julia Silge) when my friend and colleague Doug Bruster asked me if I could demo some of the methods for his honors Shakespeare course. So I put together a demo on analyzing The Taming of the Shrew in a distill website. - The work recently turned into an assignment for E 310D Intro to Digital Studies (see here).

02-27-2023

Portrait on "Into the COLAverse"

The new podcast "Into the COLAverse", hosted by Frederick Aldama, just published their latest faculty portrait, with me as guest. Listen to it here.

03-16-2022

Slides: Lexical Diversity Measures

A slide presentation for my class Digital Text Analysis, illustrating different lexical diversity statistics that are implemented in {quanteda.textstats}. More...

02-25-2022

Lecture "Why We Need Diaspora Sociolinguistics" (UWI Mona, Jamaica)

This invited lecture was delivered live via Zoom and YouTube. More...

10-26-2021

Digital Text Analysis (Spring 2022)

This will be a graduate course focused on the use of R to analyze textual data, with use cases from a variety of academic disciplines. No prior knowledge of R or statistics required. More...

10-26-2021

NWAV 49 in Austin

The 49th meeting of the sociolinguistic conference "New Ways of Analyzing Variation", NWAV 49, has just finished. I chaired the organizing committee. There is a lot of information about the conference at its website, www.nwav49.org. And the history of the conference is discussed here.

04-26-2021

Plotting DARLA-Vowels with Tableau

A new video tutorial for instructors who are looking for an easier way to have students analyze vowel data.

01-21-2021

Inaugural Speeches by Texan Presidents

We ran some computational analyses on the inaugural speeches of US Presidents from Texas. And then we compared them to Joe Biden's inaugural speech from January 20th, 2021. Check out a summary of the findings here.

George W. Bush's second inaugural speech. (Wikimedia Commons)

01-03-2021

Handbook of English Linguistics (second edition)

The Handbook of English Linguistics (2nd edition, Wiley), which I coedited, is now shipping. Here it is at the publisher's website:

https://bit.ly/hoel-e2

The editing process for the new edition took about three years. Working with these fantastic colleagues has been a truly enriching experience!

11-05-2020

The fading of WHOM

One project that started with a graduate course in 2017 is the tracking of the increasingly antique relativizer whom over time, using COHA. With the graduate students I produced exploratory statistics. In collaboration with Axel Bohmann, I am now taking a highly multivariate look at the ways in which the receding of WHOM progresses through different genre categories (again, looking at COHA). See our progress report.

08-10-2020

Digital Studies at UT Austin

As part of the growing digital studies ecosystem at UT Austin I direct the Digital Studies Graduate Portfolio. Here is the updated info website about the program: https://liberalarts.utexas.edu/ds/graduate/ds-portfolio/

05-10-2020

Texas English on the radio

Since the beginning of 2019 I have been a regular guest on the Texas Standard, KUT's weekday news magazine. Together with host David Brown I listen to the voices of well-known Texans and discuss their linguistic style. A collection of all the episodes that have aired so far is available at the link below.

https://www.texasstandard.org/stories/categories/texan-translation/

Google Sites

Report abuse