Skip to content

riya-n/nets213-final-project

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

nets213-final-project

Project: Translate for Tots

Translate for Tots will use crowdsourcing to help translate children’s books in a variety of languages. It solves the problem of children only having access to a limited number of children’s books in their native language. If more books are translated, young kids will have more options, which could also help grow their reading skills.

There will be 2 groups in our crowd: one group that transcribes the children’s book in its original language (language requirement would only have to be monolingual in the book’s language), and the other group would translate the text (bilingual requirement). The workers will provide either transcription or translation services, depending on their status as monolingual or multilingual. We will use quality checks and also have workers overlap on tasks and then compare and aggregate to find the most common answer to ensure correctness. We will also specifically have quality check tasks where we ask workers to determine if a translation or transcription is correct. Finally, we will combine the responses that have passed our quality checks mentioned previously.

Step 0 (1 point): Select foreign language books. We are planning to use books from the International Children’s Digital Library.

Step 1 (3 points): Transcribe the children’s book. This would be done by workers who are native speakers of the book’s original language.

Step 2 (4 points): Transcription quality check. The transcription would then be checked for quality through automatic comparisons and aggregation through a different set of tasks done by workers who are fluent in the original language.

Step 3 (2 points): Run transcription through Google Translate to get a rough transcription.

Step 4 (4 points): Workers then edit the Google Translate output for a more coherent translation.

Step 5 (4 points): Translation quality check. We finalize the translation through aggregation of worker responses.

Step 6 (2 points): Reassemble the books with the translations.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%