Sign in

Matthew C. Kelley

@mckphon.bsky.social
22 followers 22 following 20 posts

Computational phonetician. Assistant professor at George Mason University. Lead developer of the Phonetics.jl package and the Mason-Alberta Phonetic Segmenter. Opinions my own, not my employer's.

PostsRepliesMedia
Matthew C. Kelley @mckphon.bsky.social · 10/10/2026
Joining the collective sigh of relief among phoneticians for getting my #ICPhS2027 papers submitted tonight 😌😮‍💨
000
Reposted by Matthew C. Kelley
AAUP @aaup.org · 18/08/2026
"Professors didn’t choose the battlefield, but we’re no longer willing to pretend we aren’t standing on one." — Todd Wolfson, AAUP President
washingtonpost.com
Opinion | What professors gained by winning with El-Sayed
The American Association of University Professors' endorsement, back to school advice and more.
211740
Matthew C. Kelley @mckphon.bsky.social · 13/08/2026
Give it a try at accent.gmu.edu!
accent.gmu.edu
The Speech Accent Archive
The speech accent archive uniformly presents a large set of speech samples from a variety of language backgrounds. Native and non-native speakers of English read the same paragraph and are carefully t...
010
Matthew C. Kelley @mckphon.bsky.social · 13/08/2026
Additionally, the source files for the Archive are now available to download through OSF under the CC BY-NC-SA 4.0 license. A page on the site gives more details. (The GIFs are still uploading at the moment, so please wait until tomorrow to download if you are interested in those!)
Screenshot of the "Download" page showing the descriptions and links to the downloads.
100
Matthew C. Kelley @mckphon.bsky.social · 13/08/2026
I am happy to share that a refreshed version of the #SpeechAccentArchive has been made public today. The search functionality has been reinstated, and many files (but not all) now have plain-text copyable transcriptions.
Screenshot of the Speech Accent Archive homepage showing the new layout.Screenshot of English speaker 1, showing the plain-text IPA transcription.
100
Matthew C. Kelley @mckphon.bsky.social · 08/05/2026
Surely, there is more to this than can fit in a thread. Yet I still can't help but wonder at what that performative act is, at what point the Constitution is considered amended.
000
Matthew C. Kelley @mckphon.bsky.social · 08/05/2026
As the VA Constitution reads: "...if the people shall approve and ratify such amendment or amendments by the majority of the electors qualified to vote for the members of the General Assembly voting thereon, such amendment or amendments shall become a part of the Constitution."
100
Matthew C. Kelley @mckphon.bsky.social · 08/05/2026
The amendment would have happened on voting day, and it seems farcical for any court to say the (now changed) Constitution is unconstitutional or illegal. They would have needed to intervene before.
100
Matthew C. Kelley @mckphon.bsky.social · 08/05/2026
If the former, no amount of court opinion or action would seem to matter after voting day to (non-lawyer) me. The votes are already cast and the tabulation is merely finding out what the action was.
100
Matthew C. Kelley @mckphon.bsky.social · 08/05/2026
I wonder at what the actual performative act is when amending the VA Constitution. Is it the actual votes cast, or is it some bureaucratic certification afterward?
110
Reposted by Matthew C. Kelley
ProfKFH @profkfh.bsky.social · 05/05/2026
“Faculty members do not surrender their rights to speak as scholars and citizens when they participate in university forums—whether in classrooms, public venues, or ceremonial settings. On the contrary, such occasions are among the most visible expressions of the university’s core commitments . . .
02312
Matthew C. Kelley @mckphon.bsky.social · 05/05/2026
The widths of the boundary regions tend to match phonetic intuitions, where more ambiguous segmentations (e.g., vowel-vowel, approximant-vowel) are wider than less ambiguous segmentations (e.g., fricative-vowel, stop-vowel).
Heatmap showing the median widths for different bisegment sequences.
000
Matthew C. Kelley @mckphon.bsky.social · 05/05/2026
New annotation formats are JSONs that list each boundary for a file and its associated widths, and a large CSV with boundaries in each row. Praat TextGrids are also output with PointTiers for the boundaries to aid in visualization.
A TextGrid in Praat showing the segmentation overlaid on a spectrogram and a waveform for the first four sounds in "scholar". The boundary region between the low-back vowel and the lateral is wider than the one between the voiceless velar stop and the low-back vowel.
100
Matthew C. Kelley @mckphon.bsky.social · 05/05/2026
New paper on a gradient approach to speech segmentation! A new feature in the Mason-Alberta Phonetic Segmenter generates gradient intervals for boundary placement in forced alignment. doi.org/10.1515/phon...
doi.org
Gradient boundaries through confidence intervals for forced alignment estimates using model ensembles
Forced alignment is a common tool to align audio with orthographic and phonetic transcriptions. Most forced alignment tools provide only point-estimates of boundaries. The present project introduces a...
110
Matthew C. Kelley @mckphon.bsky.social · 04/04/2026
Seems like NSF needs to be sued to force them to stop doing illegal things. Funny how science as a human action is always political...
000
Reposted by Matthew C. Kelley
GMU-AAUP @aaupmason.bsky.social · 15/03/2026
As the great Houston Oilers coach once said, this year we knocked on the door. Next year, we'll kick the door in! We won't stop until all faculty and graduate workers have collective bargaining rights and the protection of a union.
0235
Reposted by Matthew C. Kelley
VA-AAUP @aaupvirginia.bsky.social · 15/03/2026
All @aaupvirginia members can feel proud that they played a role in this victory for state workers in VA. Even though the GA cut us out of the bill, we never once broke ranks. We lived our values. We acted in solidarity. We centered the margins. And know this: next year we'll win CB for us, too!
A photo of six members of @aaupvirginia at a rally for collective bargaining rights for all Virginia state workers.
0219
Reposted by Matthew C. Kelley
Bethany Letiecq @bletiecq.bsky.social · 18/02/2026
Let’s go @aft.org + @aaup.org! Let’s fight for inclusion and better working conditions for VA’s higher ed workers. At @aaupmason.bsky.social, under attack by Trump-Youngkin and a rogue board, we fought for President Washington’s job—and won! All workers at Mason deserve that level of fight!
1136
Matthew C. Kelley @mckphon.bsky.social · 02/01/2026
#Gramle 1102 3/5 ⬛⬛⬛⬛🟩 ⬛⬛🟩🟩🟩 🟩🟩🟩🟩🟩
010
Reposted by Matthew C. Kelley
Prof. Nicole Holliday @mixedlinguist.bsky.social · 14/08/2025
The existence of educated, qualified black people in positions of power is one of the biggest the threats to white supremacy. It’s why the first attacks were on “DEI”. Trump has never gotten over the fact that Obama was president, and so his quest is to make sure no one black ever is again.
1527
Matthew C. Kelley @mckphon.bsky.social · 14/08/2025
Clear example that Spotify's speech synthesis can't do any modicum of code switching when it read Spanish "Palabra de mujer" as [pəˈlæbɹə dɨ mjud͡ʒɚ]...
110
Matthew C. Kelley @mckphon.bsky.social · 29/07/2025
Tell me you're thin-skinned without telling me... It's clowns all the way down at Trump's DoJ. 🤡
nytimes.com
Faculty Support of George Mason’s President Draws Federal Investigation
022
Reposted by Matthew C. Kelley
GMU-AAUP @aaupmason.bsky.social · 26/07/2025
These threadbare federal investigations have one purpose: to provide a pretext for the Trump admin, Gov. Youngkin, and their minions on the BOV to take over and turn GMU into a factory for MAGA ideology
021
Reposted by Matthew C. Kelley
GMU-AAUP @aaupmason.bsky.social · 27/07/2025
042
Matthew C. Kelley @mckphon.bsky.social · 11/04/2025
New OA paper, new data set: SingleMALD! We had 40 native speakers of Canadian English do lexical decision for over 26,000 English words and 9,000 pseudowords. Evidence for participants getting faster throughout extensive testing periods. doi.org/10.3758/s134...
Plot showing average response latency decreasing over each sessions for participants, for both words and pseudowords.
110
Matthew C. Kelley @mckphon.bsky.social · 19/12/2024
Get it!!! You're gonna do great!
010
Reposted by Matthew C. Kelley
Kirby Conrod @kirbyconrod.bsky.social · 26/11/2024
Just got an email from the zombie journal soliciting my submissions so I thought I'd remind everyone that in 2015 the entire editorial board of "Lingua" resigned and created a new independent OA journal (Glossa). Don't submit to the new Lingua. slate.com/human-intere...
slate.com
Why the Editors of a Top Linguistic Journal Resigned En Masse
This post originally appeared on Inside Higher Ed.
311135
Reposted by Matthew C. Kelley
Jen Nycz @jennycz.bsky.social · 27/11/2024
My edited volume w/Víctor Fernández-Mallat (Dialect Contact: From Speaker to Community-Based Perspectives) is out! Available via Georgetown University Press & the usual sites (academic friends, please ask your library to order a copy!) #linguistics press.georgetown.edu/Book/Dialect...
three hard cover copies of a book titled "Dialect Contact: From Speaker to Community-based Perspectives", Victor Fernandez-Mallat and Jennifer Nycz, editors
45916
Reposted by Matthew C. Kelley
Dr. Ethan Kutlu @ethankutlu.bsky.social · 21/11/2024
I am thrilled to share our new paper which investigates the role of continuous measures in linguistic diversity research. We found that children who are exposed to more linguistic diversity (e.g., language, dialect, accent) are perceptually more flexible. www.nature.com/articles/s41...
nature.com
Linguistic diversity shapes flexible speech perception in school age children - Scientific Reports
Scientific Reports - Linguistic diversity shapes flexible speech perception in school age children
7334109
Reposted by Matthew C. Kelley
Scott James Perry @sjperry.bsky.social · 17/01/2024
The second article from my dissertation is now out as my first, first-authored paper! This was a joint work with @mckphon.bsky.social and my supervisor Benjamin V. Tucker. We did a corpus analysis of The Nijmegen Corpus of Casual Spanish, and this is what we found (thread) doi.org/10.1121/10.0...
121
Reposted by Matthew C. Kelley
Jon Bakos @jonbakos.bsky.social · 06/01/2024
Enshittification is the Word of the Year! #woty2023 #lsa2024
211227
Matthew C. Kelley @mckphon.bsky.social · 10/12/2023
When it does the pandoc conversation, can you have pandoc take the -s flag? In theory, that should make the tex file standalone, which might work in Overleaf?
110
Matthew C. Kelley @mckphon.bsky.social · 01/12/2023
Excited to use this as yet another place I post my #gramle results and not network with other researchers 🤭
030