Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Yandex incorrect translation #96

Open
ankit-m opened this issue Mar 17, 2016 · 2 comments
Open

Yandex incorrect translation #96

ankit-m opened this issue Mar 17, 2016 · 2 comments

Comments

@ankit-m
Copy link
Contributor

ankit-m commented Mar 17, 2016

I have a 75% Chinese to English setting. Ideally, it shouldn't translate English to English. But as per the settings and the algorithm, it selects some words and translates. This should map the words to themselves. But words are translated wrongly by Yandex.

err_ch_eng

Role-playing is converted to role-out, were is converted to the. The error is with Yandex.
yandex_err2
yandex_err

On the other hand Google Translate does not suffer from this error.
google_correct

Off the top of my head, I think we need language detection for a page.

@ceilican
Copy link
Contributor

Interesting.

Originally (a few years ago), MTW had a "detect language" option for the source language. This option would rely on Google Translate's auto-detect language feature. But somehow it wasn't working properly. That is why I disabled the feature. But maybe things are better now. Or maybe we could implement our own language detection somehow. If you have any ideas about that, please let me know.

@ankit-m
Copy link
Contributor Author

ankit-m commented Mar 18, 2016

To detect language, a fast and simple way could be to look at the Unicode encoding of the letters and find their script/language from it.

One problem with detect language is that a page may contain many languages. To find source language would be tough in this case.

We can still use Google API, but besides being a paid service, it will add a lot of overhead time.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants