-
# Crosslingual phonological feature discrimination
Similar to existing SSLR classification probes ([Cormac English et al 2022](https://aclanthology.org/2022.sigmorphon-1.9.pdf)), we evaluate whethe…
-
from Mark Seidenberg: "add phonological properties, derive layouts of how words fill phonological space over time. very important to know. got that yet?"
interesting feature to integrate into net…
-
# Multilingual pronunciation similarity
The goal of this task is to measure whether or not the models in the benchmark encode pronunciation similarity.
This task is part of the meta-task #140.
…
-
this would more properly be called the `Phonologizer`, and it could borrow heavily from spaCy's `Morphologizer`. see for reference [Wikipedia on "distinctive features"](https://en.wikipedia.org/wiki/D…
-
* This is part 2 of changing the Morpheme phonology; #638 is part 1
* Rather than having a specific strong hand or weak hand, Morphemes may include sub-features of handshapes in their phonological sp…
-
we already have things like `C[+voiced]`
whearas `+feature` means presence and `-feature` means absence of a feature. a phoneme may be [unmarked](https://en.wikipedia.org/wiki/Unmarked) with respect…
-
Create a new `PhonologicalRule` schema. Each Phonological Rule should have the following properties:
- [ ] `name`
- [ ] `description`: a prose description of this rule
- [ ] `formula`: a string r…
-
I'm just querying the coding of Russian for feature GB192. It's currently 0 but I would have thought it should be 1. To focus on one part of the paradigm, nouns whose nominative singular end in a hard…
-
# Dialect-to-Standard Normalization
The goal of this task is to evaluate to what extent speech models encode dialectal variation, by prompting models to normalize dialectal variants of Swiss German…
-
- [ ] As a researcher I need to decide which features and regressors need to be ported
* includes feature pipeline
* includes classification
This means checking if features can be properly extrac…