As machine translation becomes available in more languages with MinT and NLLB-200 models, we want to expose the improved translation tools more broadly to facilitate the creation of content for those communities.
This ticket selects an initial set of languages that had no machine translation services available before. This ticket proposes to enable machine transition using MinT, set Content Translation as a default tool and enable the mobile experience with Section Translation.
As part of this process, we'll generate the template parameter alignments (T221211) for these languages, enable access to the tool in test wiki, and communicate with the different communities to inform them about the changes.
These are the languages selected for this task:
- Kashmiri (ks/kas_Arab) ( view request and T326541 )
- Santali (sat/sat_Olck) ( view request)
- Tumbuka (tum/tum_Latn) (view request)
- Fulah/Nigerian Fulfulde (ff/fuv_Latn)
- Kabyle (kab/kab_Latn)
- Balinese (ban/ban_Latn)
- Banjar (bjn/bjn_Latn)
- South Azerbaijani (azb/azb_Arab)
- Pangasinan (pag/pag_Latn)
- Tibetan (bo/bod_Tibt)
Steps:
- Enable MinT for the selected languages.
- Enable Section Translation on selected wikis on Test Wikipedia
- Communicate with the communities in progress:
- Enable Content Translation as default in the selected Wikipedias
- Enable Section Translation in the selected Wikipedias
- Generate template parameter alignments (T221211)
- azb
-
ban(No Fasttext vector model available) - bjn
- bo
- ff
- kab
- ks
- pag
-
sat(No Fasttext vector model available) - tum
Next group of languages to enable: T337669: Enable MinT, Content and Section Translation for a 2nd group of 10 languages previously lacking machine translation