Panphon for JS ?

[guettli]

Hi, I am coding a small application to help people improve their pronounciation (especially my disabled son).

guettli/pp: Phoneme Party: Speaking Clearly should be fun

It should run in-browser, and works not that bad.

Currently the calcutation between phonemes is not as sophiscated as panphon.

But Panphon does not work in the browser.

Before I start to rewrite (or let AI rewrite) Panphon in JS, I would like to ask you:

What would you do, if you need Panphon in the browser?

Do you think it makes sense to translate the existing code to JS?

[dmort27]

PanPhon, of course, cannot be used in a JS or WASM application (using current technology) but the algorithms for implemented feature distance metrics are quite simple. Likewise, the logic for extracting rows from a feature table to map from phonemes strings to feature matrices (a prerequisite to the distance calculations) should not be hard to implement. I cannot do this myself, but I suspect that it would be just a few hours of work for someone proficient in JS.

[guettli]

@dmort27 thank you for your feedback. I hope that I have time for that during the next weeks. I will post here, when there is something usable.