Page MenuHomePhabricator

Accumulate lexemes without pronunciations and offer to feed them into a Lingua Libre session
Open, LowPublicFeature

Description

Feature summary:
It would be neat if while illustrating lexemes using Luthor, the tool would notify the user of missing pronunciations and optionally add them to a list (kept in session), which at any point the user can choose to use to initiate a pronunciation recording session in the Lingua Libre tool.

(this assumes there is a way to initiate such a session with a given word-list in LL.)

Benefits (why should this be implemented?):
encourage pronunciation recordings and create added-value for each lexeme processed in Luthor.

Event Timeline

Yug triaged this task as Low priority.Jul 13 2025, 9:05 AM

That seems like a fantastic idea. I would love to use such functionality.

I'm pretty sure there is a way to pass a list into LL. There is right now, @Yug : will it still possible in the next version of LL?
@Ijon technically, it is forms representation not lexeme (this is actually true for both examples in Luthor and recording in LL), one trivial solution is to take the form corresponding to the lemma but maybe there is an opportunity to do something better to do here.

The new version is in beta test on https://lingua-libre.toolforge.org . In the List step, you can open the Wikidata generator, which expects a wikidata query url. You can feed it with a « lexemes in language xxx without pronunciations », and it willl load your lexemes. It s a bit complex but it already works. Just ask a GPT for that sparql and go to query.wikidata.org, run it, to get tgat query url.

Alternatively, i could add a much simplier button with « Lexemes in xxx without records », to feef the beast.

That is a pretty good idea.