GiellaLT

GiellaLT provides an infrastructure for rule-based language technology aimed at minority and indigenous languages, and streamlines building anything from keyboards to speech technology. Read more about Why. See also How to get started, and our Privacy document.

How to use the morphological parsers

You must have set up the infrastructure environment for this to work. If you did not, look at the Getting Started page under the Overview section on the frontpage, (here). For analysing when you don’t have this infrastructure setup, see this page.

Analysing and generating

… with some handy aliases:

Aliases for word analysis and generation

To analyse a word, write hu + your language code: husme or hufin or… follwed by ENTER. Then write the word you want to analyse, and press ENTER. The command also reads input, one word at a time:

echo gillii | husme
cat wordlist.txt | husme

To generate, use hd + the language code, where input should look like the previous output, e.g. echo giella+N+Sg+Ill | hdsme

Aliases for sentence analysis

cat text | dis.sh -l sme
cat text | dep.sh -l sme

will give a disambiguated or dependency analysis of any language in our infra (replace sme with your language). There are some other options as well, check with dis.sh -h.

… with the commands themselves:

We assume you stand in the lang-XXX folder, where XXX is the 3-letter code of your language (sme for North Saami, etc.).

  1. In order to analyse words one at a time, write
    • hfst-lookup -q src/analyser-gt-desc.hfstol
    • (followed by ENTER)
  1. For generation, write
    • hfst-lookup -q src/generator-gt-desc.hfstol
    • (followed by ENTER)
  1. For testing, you may also write a file with one wordform on each line, and then feed that to the analyser (example here is with a file testfile.txt):

cat testfile.txt | hfst-lookup -q src/analyser-gt-desc.hfstol | less

Text analysis (hfst)

The command to analyse text is:

cat testfile.txt | hfst-tokenise --giella-cg tools/tokenisers/tokeniser-disamb-gt-desc.pmhfst

In case the transducer contains weights, the constraint grammar may make use of them, as follows

cat text | hfst-tokenise --giella-cg tools/tokenisers/tokeniser-disamb-gt-desc.pmhfst | ...

Please note that the file tools/tokenisers/tokeniser-disamb-gt-desc.pmhfst is not built by default. To enable building it, configure as follows:

./configure --enable-tokenisers

Manipulating the output

Instead of just showing the result on the screen as running text (as above), much can be done to manipulate it. Here are some examples, all the textstrings should be added after the dis.sh etc. above.

| grep '+N+Pl' > plnouns
(to get all plural nouns and save them to the file plnouns)

| grep -v '\?' | cut -f2 | sort | uniq -c | sort -nr | less
(to get a frequency list of the lexemes that the parser recognizes.

| grep '\?' | sort | uniq -c | sort -nr | less
(to get a frequency list of the words that the parser does not recognize)

| grep '\+\?' | sort | uniq -c | sort -nr | less
(to get a frequency list of the word forms that the parser does not recognize)

Old, obsolete documentation

Aliases for text analysis (xfst and hfst)

You may have a family of aliases set up on your machine. Find out if you have by writing alias smedis. If the answer is sent-proc.sh -s dis, they are set up. If the answer is -bash: alias: smedis: not found, they are not.

The aliases contain a pipeline combining perl pre- and postprocessing with xfst transducers and constraint grammar. These aliases may be written anywhere (replace “sme” with your own language code). Note that they need the xfst compiler.

These aliases may be used in two ways: either write the alias followed by a sentence in quotes

smedis "Mun lean boahtán."

Or, alternatively, pipe a file through it:

`cat testfile.txt | smedis``

Sitemap