Download for free from Play Store!
Or try first a web version of the translator
The Human Language Compiler differs from other translation systems that are currently available, by using grammars and semantics to perform the translation. Other systems use either statistics or shallow rules based on dictionary lookup. Our technology is similar to the way in which a compiler translates program code to machine code - hence the name of our system.
The technology enables the system to be compact in size and give control on quality. It indicates confidence with colours:
You can translate both speech and text, as selected in the menu in the upper right corner. Translation works between any pair of the supported languages, which means 110 language pairs at the moment. But different languages are on different levels of development. The following table gives a rough idea of what to expect:
| coverage | quality | speed | speech | |
|---|---|---|---|---|
| Bulgarian | in only | |||
| Chinese | diff in/out | |||
| Dutch | ||||
| English | ||||
| Finnish | ||||
| French | ||||
| German | ||||
| Hindi | out only | |||
| Italian | ||||
| Spanish | ||||
| Swedish |
When you tap on a translation you get a screen with alternative translations. Tapping on each of the alternatives gives you grammatical information: an inflection table, if it is a single word, and a syntax tree otherwise.
The app also provides an input method which you can use as an alternative keyboard which allows you to do translation from other applications, for instance while you are entering SMS or e-mail. To activate it go to Settings > Language & input.
The Human Languge Compiler works completely off-line, without internet connection, although being on-line may give you better speech input and output and enable it for more languages. The app is small (around 24 MB for all languages). We do not show ads.
Powered by GF, built by support from the GF community and from Digital Grammars.
Digital Grammars can tailor this app to you needs - tell us what you want to see in the green area!
Publication: EACL 2014 demo paper