What are parallel documents?
Parallel documents are pairs of documents where one is the translation of the other. One document in the pair contains sentences in the source language and the other document contains these sentences translated into the target language. It doesn’t matter which language is marked as “source” and which language is marked as “target” – a parallel document can be used to train a translation system in either direction.
Use of parallel documents
Parallel documents are used by the system:
To learn how words, phrases and sentences are commonly mapped between the two languages.
To learn how to process the appropriate context depending on the surrounding phrases. A word may not always translate to the exact same word in the other language.
As a best practice, make sure that there is a 1:1 sentence correspondence between the source and target language versions of the documents.
Documents uploaded are private to each workspace and can be used in as many projects or trainings as you like. Sentences extracted from your documents are stored separately in your repository as plain Unicode text files and are available for you delete. Do not use the Custom Translator as a document repository, you will not be able to download the documents you uploaded in the format you uploaded them.
If your project is domain (category) specific, your documents should be consistent in terminology within that category. The quality of the resulting translation system depends on the number of sentences in your document set and the quality of the sentences. The more examples your documents contain with diverse usages for a word specific to your category, the better job the system can do during translation.
You will need a minimum of 10,000 parallel sentences to train a system. As a best practice, you can continuously add more parallel content and retrain, to improve the quality of your translation system.
- Learn how to use a dictionary in Custom Translator.
We'd love to hear your thoughts. Choose the type you'd like to provide:
Our feedback system is built on GitHub Issues. Read more on our blog.