1
0
forked from GitHub/gf-core

started demo pages for translation and resource grammars

This commit is contained in:
aarne
2014-01-10 14:54:21 +00:00
parent ac33c8a851
commit 8b85e07561
4 changed files with 193 additions and 0 deletions

View File

@@ -0,0 +1,35 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML>
<HEAD>
<META NAME="generator" CONTENT="http://txt2tags.org">
<TITLE>Computational Grammars for the World</TITLE>
</HEAD><BODY BGCOLOR="white" TEXT="black">
<CENTER>
<H1>Computational Grammars for the World</H1>
<FONT SIZE="4"><I>The GF Resource Grammar Library</I></FONT><BR>
</CENTER>
<P>
Did you ever struggle with the inflection of French verbs? Or with getting the words in a German sentence into the right order?
Don't worry: these problems are as easy for a computer as to calculate 2+2. This page will give you access to computer programs that
</P>
<UL>
<LI>help you construct sentences and other expressions
<LI>help you inflect words
<LI>help you covert sentences to different tenses, and other similar operations
<LI>help you compare languages by translating them and linking corresponding parts
<LI>allow you to practice your language skills by automatically generated and graded exercises
<LI>explain grammar rules in a mathematically precise and shortest possible manner
</UL>
<P>
<I>Coming soon!</I> for 29 languages.
</P>
<P>
Powered by <A HREF="..">GF</A> and its <A HREF="../lib/doc/synopsis.html">Resource Grammar Library</A>
</P>
<!-- html code generated by txt2tags 2.6 (http://txt2tags.org) -->
<!-- cmdline: txt2tags -thtml resourcegrammars.t2t -->
</BODY></HTML>

View File

@@ -0,0 +1,19 @@
Computational Grammars for the World
The GF Resource Grammar Library
Did you ever struggle with the inflection of French verbs? Or with getting the words in a German sentence into the right order?
Don't worry: these problems are as easy for a computer as to calculate 2+2. This page will give you access to computer programs that
- help you construct sentences and other expressions
- help you inflect words
- help you covert sentences to different tenses, and other similar operations
- help you compare languages by translating them and linking corresponding parts
- allow you to practice your language skills by automatically generated and graded exercises
- explain grammar rules in a mathematically precise and shortest possible manner
//Coming soon!// for 29 languages.
Powered by [GF ..] and its [Resource Grammar Library ../lib/doc/synopsis.html]

77
demos/translation.html Normal file
View File

@@ -0,0 +1,77 @@
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML>
<HEAD>
<META NAME="generator" CONTENT="http://txt2tags.org">
<TITLE>Translation with GF</TITLE>
</HEAD><BODY BGCOLOR="white" TEXT="black">
<CENTER>
<H1>Translation with GF</H1>
<FONT SIZE="4"><I>Powered by multilingual grammars</I></FONT><BR>
</CENTER>
<H1>Try it online</H1>
<P>
<B>Coming soon</B>:
<I>Here you will be able to select your source and target language and translate whatever you want.</I>
</P>
<H1>Download an app</H1>
<P>
<B>Coming soon</B>: <I>link to an Android app, with a barcode.</I>
</P>
<H1>What you can expect</H1>
<P>
The program is designed to translate anything, just like Google translate and Bing. It has some advantages over them:
</P>
<UL>
<LI>It uses <B>grammars</B> (as opposed to statistics), which makes it more correct in many cases.
<LI>It is <B>open source</B>, which means you can inspect, copy, and modify it in what ever way you want.
<LI>It is <B>light weight</B>, so that you can run it on your own computer and even on a smartphone.
<LI>It gives you <B>linguistic information</B>, such as word inflections.
<LI>It gives you <B>confidence information</B>, warning about translations that it is not so sure about.
</UL>
<P>
However, this is not a mature product but a research prototype. It therefore has some shortcomings:
</P>
<UL>
<LI>It has <B>limited coverage</B>, which means that it misses many words and syntactic constructions.
<LI>It can be <B>clumsy</B>, for instance when it doesn't detect an idiom but translates too literally.
<LI>It can be <B>slow</B>, in particular with long sentences and with some input languages.
</UL>
<P>
Here are some hints to use it in a satisfactory way:
</P>
<UL>
<LI>Start with short sentences (less than ten words), preferably from English or Swedish, to get an idea of the best performance.
<LI>Take a look at the alternative translations, in particular for sentences with low confidence level.
<LI>Send us feedback, and consider contributing to the improvement of the system!
</UL>
<H1>How it works</H1>
<P>
Powered by <A HREF="../">Grammatical Framework</A>
</P>
<P>
Publications explaining the translation method:
</P>
<H1>Get the sources</H1>
<P>
Download from <A HREF="../download/">here</A>
</P>
<!-- html code generated by txt2tags 2.6 (http://txt2tags.org) -->
<!-- cmdline: txt2tags -thtml translation.t2t -->
</BODY></HTML>

62
demos/translation.t2t Normal file
View File

@@ -0,0 +1,62 @@
Translation with GF
Powered by multilingual grammars
=Try it online=
**Coming soon**:
//Here you will be able to select your source and target language and translate whatever you want.//
=Download an app=
**Coming soon**: //link to an Android app, with a barcode.//
=What you can expect=
The program is designed to translate anything, just like Google translate and Bing. It has some advantages over them:
- It uses **grammars** (as opposed to statistics), which makes it more correct in many cases.
- It is **open source**, which means you can inspect, copy, and modify it in what ever way you want.
- It is **light weight**, so that you can run it on your own computer and even on a smartphone.
- It gives you **linguistic information**, such as word inflections.
- It gives you **confidence information**, warning about translations that it is not so sure about.
However, this is not a mature product but a research prototype. It therefore has some shortcomings:
- It has **limited coverage**, which means that it misses many words and syntactic constructions.
- It can be **clumsy**, for instance when it doesn't detect an idiom but translates too literally.
- It can be **slow**, in particular with long sentences and with some input languages.
Here are some hints to use it in a satisfactory way:
- Start with short sentences (less than ten words), preferably from English or Swedish, to get an idea of the best performance.
- Take a look at the alternative translations, in particular for sentences with low confidence level.
- Send us feedback, and consider contributing to the improvement of the system!
=How it works=
Powered by [Grammatical Framework ../]
Publications explaining the translation method:
=Get the sources=
Download from [here ../download/]
=Send us feedback=