Enable Softcatalà models for more language pairs in MinT test instance
Closed, ResolvedPublic
Actions

Assigned To

Authored By

	Pginer-WMF
	Jan 18 2024, 11:01 AM

Project Tags

Referenced Files

	F44622067: translate.wmcloud.org_(Wiki Tablet) (17).png
	Apr 5 2024, 9:06 AM

	F44622038: translate.wmcloud.org_(Wiki Tablet) (16).png
	Apr 5 2024, 9:06 AM

	F44621826: translate.wmcloud.org_(Wiki Tablet) (15).png
	Apr 5 2024, 9:06 AM

	F44621612: translate.wmcloud.org_(Wiki Tablet) (14).png
	Apr 5 2024, 9:06 AM

	F44618628: translate.wmcloud.org_(Wiki Tablet) (12).png
	Apr 5 2024, 9:06 AM

	F44618161: translate.wmcloud.org_(Wiki Tablet) (13).png
	Apr 5 2024, 9:06 AM

	F44620682: translate.wmcloud.org_(Wiki Tablet) (11).png
	Apr 5 2024, 9:06 AM

	F44620570: translate.wmcloud.org_(Wiki Tablet) (10).png
	Apr 5 2024, 9:06 AM

Subscribers

View All 12 Subscribers

Description

The MinT test instance supports now selecting which model to use when multiple are available (T338608). Enabling more language pairs supported by the Softcatalà models (T284905) will allow communities to compare whether they work better than the current alternatives.

This ticket proposes to enable Softcatalà models for the supported language pairs (as non-default for now):

German (de) ↔ Catalan (ca)
French (fr) ↔ Catalan (ca)
Galician (gl) ↔ Catalan (ca)
Italian (it) ↔ Catalan (ca)
Japanese (ja) ↔ Catalan (ca)
Dutch (nl) ↔ Catalan (ca)
Occitan (oc) ↔ Catalan (ca)
Portuguese (pt) ↔ Catalan (ca)
Spanish (es) ↔ Catalan (ca)
~~English (en) ↔ Catalan (ca)~~ (already enabled and default)

Details

Related Changes in Gerrit:

Subject	Author	Repo	Branch	Lines +/-
Update MinT to 2024-08-05-062247-production	KartikMistry	operations/deployment-charts	master	+1 -1
Softcatalà: Add missing config for ca<->es pair	KartikMistry	mediawiki/services/machinetranslation	master	+5 -1
config/Softcatalà: Fix typo ti->it	KartikMistry	mediawiki/services/machinetranslation	master	+1 -1
Update MinT to 2024-03-26-120044-production	KartikMistry	operations/deployment-charts	master	+1 -1
Enable Softcatalà models for more language pairs	Santhosh	mediawiki/services/machinetranslation	master	+106 -6

Customize query in gerrit

Related Objects
Search...

Status	Assigned	Task
Open	None	T76456 Language Engineering tracker of trackers (tracking)
Invalid	None	T76454 [Epic] ContentTranslation - MT support
Open	None	T86700 Add more machine translation services (tracking)
Resolved	Pginer-WMF	T284905 Softcatalà translator - requested for integration as an MT service for CX
Resolved	KartikMistry	T355304 Enable Softcatalà models for more language pairs in MinT test instance

Event Timeline

Pginer-WMF created this task.Jan 18 2024, 11:01 AM

Pginer-WMF triaged this task as Medium priority.

Pginer-WMF moved this task from Backlog to Development environment on the MinT board.Feb 5 2024, 6:15 PM

Pginer-WMF moved this task from Development environment to Adding languages on the MinT board.

Change #1014431 had a related patch set uploaded (by Santhosh; author: Santhosh):

[mediawiki/services/machinetranslation@master] Enable Softcatalà models for more language pairs

https://gerrit.wikimedia.org/r/1014431

gerritbot added a project: Patch-For-Review.Mar 26 2024, 9:07 AM

santhosh claimed this task.Mar 26 2024, 9:29 AM

santhosh moved this task from Quarter Backlog to In Review on the Language-Team (Language-2024-January-March) board.

Change #1014431 merged by jenkins-bot:

[mediawiki/services/machinetranslation@master] Enable Softcatalà models for more language pairs

https://gerrit.wikimedia.org/r/1014431

Change #1014729 had a related patch set uploaded (by KartikMistry; author: KartikMistry):

[operations/deployment-charts@master] Update MinT to 2024-03-26-120044-production

https://gerrit.wikimedia.org/r/1014729

Change #1014729 merged by jenkins-bot:

[operations/deployment-charts@master] Update MinT to 2024-03-26-120044-production

https://gerrit.wikimedia.org/r/1014729

Mentioned in SAL (#wikimedia-operations) [2024-03-27T07:11:01Z] <kart_> Updated MinT to 2024-03-26-120044-production (T347930, T355304, T349487)

Stashbot mentioned this in T347930: Odia Language Translation Number not translating.Mar 27 2024, 7:11 AM

Stashbot mentioned this in T349487: Improve MinT punctuation support for Japanese.

santhosh moved this task from In Review to Needs QA on the Language-Team (Language-2024-January-March) board.Mar 28 2024, 5:26 AM

Nikerabbit removed a project: Patch-For-Review.Apr 2 2024, 11:18 AM

Pginer-WMF edited projects, added: Language-Team (Language-2024-April-June); removed: Language-Team (Language-2024-January-March).Apr 2 2024, 3:19 PM

Pginer-WMF moved this task from Quarter Backlog to Needs QA on the Language-Team (Language-2024-April-June) board.

@KartikMistry. I reviewed the enabled pairs, the status are shown in a table below with issues for Italian, Occitan, Spanish and, maybe, Japanese. The test instance went down and I could not complete testing a couple combinations (see the notes, and double check as the other issues are fixed):

Language pair	Status	Notes
German (de) ↔ Catalan (ca)	✅
French (fr) ↔ Catalan (ca)	✅
Galician (gl) ↔ Catalan (ca)	✅
Italian (it) → Catalan (ca)	❌	Translation attempts result in an error. The model is listed as softcatala-ti-ca, which seems to use `ti` for the Italian code while the correct code for Italian is `it` instead. So there may be a typo somewhere. The reverse direction for the language pair (i.e, Catalan (ca) → Italian (it) works as expected.
Japanese (ja) ↔ Catalan (ca)	❓	The models provides a translation but it often has some issues with markup (notice the "amp;apos;" in the translated content. We may want to check if this is just noise introduced by the model, or due to some processing; and decide whether a different model such as NLLB-200 may be a better default for this pair.
Dutch (nl) ↔ Catalan (ca)	✅
Occitan (oc) → Catalan (ca)	❌	Softcatalà model is not available, only NLLB-200 is listed. The reverse direction for the language pair (i.e, Catalan (ca) → Occitan (oc) works as expected.
Portuguese (pt) ↔ Catalan (ca)	✅
Spanish (es) → Catalan (ca)	❌	Softcatalà model is not available, only NLLB-200 is listed. The reverse direction for the language pair (i.e, Catalan (ca) → Spanish (es) works as expected.

Nikerabbit removed santhosh as the assignee of this task.May 13 2024, 12:19 PM

Nikerabbit moved this task from In Progress to Priority: Translation on the Language-Team (Language-2024-April-June) board.

Pginer-WMF assigned this task to KartikMistry.Jun 6 2024, 9:42 AM

KartikMistry edited projects, added: LPL Essential (LPL Essential 2024 Jul-Oct); removed: Language-Team (Language-2024-April-June).Jul 2 2024, 6:46 AM

I made some tests for the es ↔ ca pairs and the quality seems better than NLLB-200. We can consider setting it as the default model for MinT on this language pair.

Pginer-WMF mentioned this in T370749: Re-run the MT service usage report (2024).Jul 23 2024, 8:32 AM

Change #1056401 had a related patch set uploaded (by KartikMistry; author: KartikMistry):

[mediawiki/services/machinetranslation@master] config/Softcatalà: Fix typo ti->it

https://gerrit.wikimedia.org/r/1056401

gerritbot added a project: Patch-For-Review.Jul 24 2024, 6:54 AM

Change #1056403 had a related patch set uploaded (by KartikMistry; author: KartikMistry):

[mediawiki/services/machinetranslation@master] Softcatalà: Add missing config for ca<->es pair

https://gerrit.wikimedia.org/r/1056403

Change #1056401 merged by jenkins-bot:

[mediawiki/services/machinetranslation@master] config/Softcatalà: Fix typo ti->it

https://gerrit.wikimedia.org/r/1056401

Change #1056403 merged by jenkins-bot:

[mediawiki/services/machinetranslation@master] Softcatalà: Add missing config for ca<->es pair

https://gerrit.wikimedia.org/r/1056403

Maintenance_bot removed a project: Patch-For-Review.Jul 24 2024, 3:31 PM

@Pginer-WMF I have updated test instance https://translate.wmcloud.org/ and issues related to Italian (it) → Catalan (ca) and es ↔ ca pairs should be fixed. Let me know.

In T355304#10013670, @KartikMistry wrote:

@Pginer-WMF I have updated test instance https://translate.wmcloud.org/ and issues related to Italian (it) → Catalan (ca) and es ↔ ca pairs should be fixed. Let me know.

Thanks Kartik. This is working now.

Change #1057421 had a related patch set uploaded (by KartikMistry; author: KartikMistry):

[operations/deployment-charts@master] Update MinT to 2024-07-24-145137-production

https://gerrit.wikimedia.org/r/1057421

gerritbot added a project: Patch-For-Review.Jul 29 2024, 3:54 AM

Change #1057421 merged by jenkins-bot:

[operations/deployment-charts@master] Update MinT to 2024-08-05-062247-production

https://gerrit.wikimedia.org/r/1057421

Maintenance_bot removed a project: Patch-For-Review.Aug 6 2024, 7:31 AM

Mentioned in SAL (#wikimedia-operations) [2024-08-06T07:39:57Z] <kart_> Updated MinT to 2024-08-05-062247-production (T363308, T355304, T368521)

Stashbot mentioned this in T368521: MinT leaks template metadata in translated content.Aug 6 2024, 7:39 AM

Stashbot mentioned this in T363308: Server-side caching for MinT.

Enable Softcatalà models for more language pairs in MinT test instanceClosed, ResolvedPublicActions

Description

Details

Related ObjectsSearch...

Event Timeline

Enable Softcatalà models for more language pairs in MinT test instance
Closed, ResolvedPublic
Actions

Related Objects
Search...