"It would be nice to add pronunciation for all words (or add a phonetic transcription in phonetic international alphabet or both), as I noticed that some words are not accompanied by a recording."
[ ] Phonetic transcription? - In training the sparse transcription model, we're using that package developed by Eddie and other NRC folks. That could be used to generate phonetic representations that could be presented on the details page (as an option).
[ ] For now I'm a bit wary about generating synthesized audio for all of the itwêwina content. If we'd do that, I'd restrict it to a few chosen, diagnostically useful inflected word forms for the entire DB content, and then fuller paradigms for some select lemmas representing various inflectional patterns.
"It would be nice to add pronunciation for all words (or add a phonetic transcription in phonetic international alphabet or both), as I noticed that some words are not accompanied by a recording."