S A P O (Sound Adaptation & Phonetic Orchestration) is an open-source project that aims at building a cross-language database that maps how every word sounds in every other language.
To generate this dataset, SAPO uses state-of-the-art AI models, which require ongoing funding. For now, this funding comes from the SAPO mobile and web apps, which provide phonetic transliteration and semantic translation. Users subscribe with a monthly token budget, and any unused portion is redirected to background processes that expand and improve the open phonetic database.
Currently, there is only an iOS app available on the app store: https://apps.apple.com/us/app/s-a-p-o-transl-ate-iterate/id6752793342
An Android and Web-app will be coming soon.