Wikisource OCR: Accept Google options on the API
Closed, ResolvedPublic3 Estimated Story Points
Actions

Assigned To

Authored By

	HMonroy
	Apr 15 2021, 2:39 AM

Description

Acceptance Criteria:

The new Wikimedia OCR should accept Google options through the API like: multiple languages.

Related Objects

Mentioned In: T279405: Add on-wiki UI for selecting languages
T280213: Wikisource OCR: Accept Tesseract options on the API
Mentioned Here: T282073: Add API endpoint to retrieve supported languages

Event Timeline

HMonroy created this task.Apr 15 2021, 2:39 AM

Restricted Application added a subscriber: • Aklapper. · View Herald TranscriptApr 15 2021, 2:39 AM

ldelench_wmf set the point value for this task to 3.Apr 15 2021, 11:25 PM

Samwilson claimed this task.Apr 16 2021, 4:58 AM

Samwilson moved this task from Ready 🎬 to In Development 💻 on the Community-Tech (CommTech-Sprint-1) board.

PR: https://github.com/wikimedia/wikimedia-ocr/pull/16

Samwilson mentioned this in T280213: Wikisource OCR: Accept Tesseract options on the API.Apr 16 2021, 6:35 AM

HMonroy moved this task from Review/Feedback 💬 to QA 🐛 on the Community-Tech (CommTech-Sprint-1) board.Apr 27 2021, 10:52 PM

QA notes: In the web interface, currently you can only enter multiple langs via the URL, i.e. https://ocr-test.wmcloud.org/?langs[]=en&langs[]=fr

Follow-up PR that makes language input in the UI a bit more intuitive: https://github.com/wikimedia/wikimedia-ocr/pull/21

jayantanth subscribed.Apr 29 2021, 10:54 AM

You can see the effect of passing multiple languages here in the returned text:

You can see a slight change in the UI, where the language inputs are a list:

I guess this is so we can have multiple language inputs, e.g.:

Better UI is merged (single input with tag-style language codes). Live on the test site (only, so far). @dom_walden, not sure if you want to have another look at this.

Samwilson mentioned this in T279405: Add on-wiki UI for selecting languages.May 3 2021, 7:20 AM

In T280214#7051507, @Samwilson wrote:

Better UI is merged (single input with tag-style language codes). Live on the test site (only, so far). @dom_walden, not sure if you want to have another look at this.

Thanks. This looks good to me.

When entering languages there is no autocomplete, the user has to know the correct name for the language.

There doesn't seem to be a limit to how many languages you can enter.

When entering languages there is no autocomplete, the user has to know the correct name for the language.

Partially addressed in T282073: Add API endpoint to retrieve supported languages, see comment at T282073#7064537

• NRodriguez moved this task from Product sign-off 🤘 to Done 🏁 on the Community-Tech (CommTech-Sprint-1) board.May 24 2021, 5:12 PM

ldelench_wmf closed this task as Resolved.May 24 2021, 7:32 PM

	F34441407: language_input.png
	May 5 2021, 8:37 AM

	F34435567: multiple_ui_mult.png
	Apr 30 2021, 1:25 PM

	F34435565: multiple_ui.png
	Apr 30 2021, 1:25 PM

Wikisource OCR: Accept Google options on the APIClosed, ResolvedPublic3 Estimated Story PointsActions

Description

Related Objects

Event Timeline

Wikisource OCR: Accept Google options on the API
Closed, ResolvedPublic3 Estimated Story Points
Actions