Apertium, Tvorba novej dvojice jazykov

Apertium - Tvorba novej dvojice jazykov

Tento dokument popíše, ako vytvorit novú dvojicu jazykov pre Apertium, systém pre strojový preklad, od začiatku. This HOWTO document will describe how to start a new language pair for the Apertium machine translation system from scratch.

Nepredpokladá žiadne znalosti ohľadom ligvistiky alebo strojového prekladu okrem schopnosti rozlišovať slovné druhy. It does not assume any knowledge of linguistics, or machine translation above the level of being able to distinguish nouns from verbs (and prepositions etc.)

Introduction

Apertium je, ako ste si už pravdepodobne uvedomili, systém pre strojový preklad. Nie je to však presné, ide o platformu pre strojový preklad. Poskytuje podklad a nástroje ktoré vám umožnia vybudovať si takýto systém. Jediné, čo musíte robiť, je poskytnúť mu dáta, obsahujúce trojicu slovníkov a pravidlá pre poradie slov a gramatiku.

Apertium is, as you've probably realised by now, a machine translation system. Well, not quite, it's a machine translation platform. It provides an engine and toolbox that allow you to build your own machine translation systems. The only thing you need to do is write the data. The data consists, on a basic level, of three dictionaries and a few rules (to deal with word re-ordering and other grammatical stuff).

Pre podrobnejší úvod do fungovania celého systému, na stránke Publications nájdete výborné zdroje. For a more detailed introduction into how it all works, there are some excellent papers on the Publications page..

You will need

• lttoolbox (>= 3.0.0)
• libxml utils (xmllint etc.)
• apertium (>= 3.0.0)
• textový editor (alebo XML editor)

• lttoolbox (>= 3.0.0)
• libxml utils (xmllint etc.)
• apertium (>= 3.0.0)
• a text editor (or a specialised XML editor if you prefer)

Inštaláciu týchto programov popisuje dokumentácia projektu Apertium. This document will not describe how to install these packages, for more information please see the documentation section of the Apertium website.

What does a language pair consist of?

Apertium je shallow-transfer systém pre strojový preklad. To znamená, že pracuje zo slovníkmi a pravidlami na preklad, na rozdiel od deep-transfer systémov, ktoré vykonávajú plnú syntaktickú analýzu. Pravidlá obyčajne popisujú operácie nad skupinami lexikálnych jednotiek, nie nad syntaktickými stromami. V podstate obsahuje trojicu slovníkov:

1. Morfologický slovník pre jazyk xx obsahuje pravidlá pre skloňovanie a časovanie slov v jazyku xx. V našom príklade sa bude nazývať apertium-sh-en.sh.dix
2. Morfologický slovník pre jazyk yy obsahuje pravidlá pre skloňovanie a časovanie slov v jazyku yy. V našom príklade sa bude nazývať code>apertium-sh-en.en.dix
3. Dvojjazyčný slovník obsahuje spojenia medzi slovami a symbolmi oboch jazykov. V našom príklade sa bude nazývať apertium-sh-en.sh-en.dix

Apertium is a shallow-transfer type machine translation system. Thus, it basically works on dictionaries and shallow transfer rules. In operation, shallow-transfer is distinguished from deep-transfer in that it doesn't do full syntactic parsing, the rules are typically operations on groups of lexical units, rather than operations on parse trees. At a basic level, there are three main dictionaries:

1. The morphological dictionary for language xx: this contains the rules of how words in language xx are inflected. In our example this will be called: apertium-sh-en.sh.dix
2. The morphological dictionary for language yy: this contains the rules of how words in language yy are inflected. In our example this will be called: apertium-sh-en.en.dix
3. Bilingual dictionary: contains correspondences between words and symbols in the two languages. In our example this will be called: apertium-sh-en.sh-en.dix

Oba jazyky vo dvojici môžu byť rovnako cieľ aj zdroj prekladu. In a translation pair, both languages can be either source or target for translation, these are relative terms.

Taktiež existujú dva súbory pre pravidlá prepisu. Tieto pravidlá riadia poradie slov vo vete, ako napr. chat noir -> cat black -> black cat. Naviac riadia zhody v rode, čísle atď. Tieto pravidlá môžu vkladať alebo mazať lexikálne položky, ako bude popísané neskôr. Tieto súbory sú:

There are also two files for transfer rules. These are the rules that govern how words are re-ordered in sentences, e.g. chat noir -> cat black -> black cat. It also governs agreement of gender, number etc. The rules can also be used to insert or delete lexical items, as will be described later. These files are:

• pravidlá pre prepis z jazyka xx do jazyka yy, popisujúce zmeny jazyka xx pri preklade do jazyka yy. V našom príklade budú v súbore apertium-sh-en.sh-en.t1x
• pravidlá pre prepis z jazyka yy do jazyka xx, popisujúce zmeny jazyka yy pri preklade do jazyka xx. V našom príklade budú v súbore apertium-sh-en.en-sh.t1x

• language xx to language yy transfer rules: this file contains rules for how language xx will be changed into language yy. In our example this will be: apertium-sh-en.sh-en.t1x
• language yy to xx language transfer rules: this file contains rules for how language yy will be changed into language xx. In our example this will be: apertium-sh-en.en-sh.t1x

Veľké množstvo jazykových dvojíc má aj ďalšie súbory, no tu sa nimi nebudeme zaoberať. Vyššie popísané súbory stačia na vytvorenie funkčného systému. Many of the language pairs currently available have other files, but we won't cover them here. These files are the only ones required to generate a functional system.

Language pair

Ako ste už možno zistili z názvov súborov, tento návod použije ako príklad Srbochorváčtiny do Angličtiny, aby vysvetlil spôsob tvorby základného systému. Nie je to ideálna dvojica, pretože systém pracuje lepšie pre príbuzné jazyky, no nebude to problém pre jednoduché príklady použité v dokumente. As may have been alluded to by the file names, this HOWTO will use the example of translating Serbo-Croatian to English to explain how to create a basic system. This is not an ideal pair, since the system works better for more closely related languages. This shouldn't present a problem for the simple examples given here.

A brief note on terms

Je zopár pojmov, ktoré treba chápať pre tvorbu novej dvojice. There are number of terms that will need to be understood before we continue.

Prvý je lemma. Lemma je základný tvar slova, ako býva uvedená v slovníkoch. Napríklad lemma slova mačkami je mačka. V Angličtine to bude väčšinou jednotné číslo podstatného mena, alebo neurčitok slovesa bez to. The first is lemma. A lemma is the citation form of a word. It is the word stripped of any grammatical information. For example, the lemma of the word cats is cat. In English nouns this will typically be the singular form of the word in question. For verbs, the lemma is the infinitive stripped of to, e.g. the lemma of was would be be.

Ďalší pojem je symbol. V kontexte systému Apertium, symbol je gramatická vlastnosť. Slovo mačky je množné číslo podstatného mena, preto bude mať symbly podstatné meno a plurál. Na vstupe a výstupe modulov pre Apertium, symboly bývajú väčšinou určené medzi ostrými zátvorkami: The second is symbol. In the context of the Apertium system, symbol refers to a grammatical label. The word cats is a plural noun, therefore it will have the noun symbol and the plural symbol. In the input and output of Apertium modules these are typically given between angle brackets, as follows:

• <n>; for noun.
• <pl>; for plural.

Iné príklady symbolov sú <sg> singular, <p1> prvá osoba, atď. Napísané v ostrých zátvorkách, symboly môžu byť označované ako tagy. Je vhodné poznamenať, že v časti dvojíc jazykov sú menjá symbolov vytvorené skrátením Katalánčiny, ako napríklad vbhaver, vb (verb) a haver ("mať" v Katalánčine). Symboly sú definované v <sdef> tagoch a používane v <s> tagoch. Other examples of symbols are <sg> singular, <p1> first person, <pri> present indicative, etc. When written in angle brackets, the symbols may also be referred to as tags. It is worth noting that in many of the currently available language pairs the symbol definitions are acronyms or contractions of words in Catalan. For example, vbhaver — from vb (verb) and haver ("to have" in Catalan). Symbols are defined in <sdef> tags and used in <s> tags.

Tretí výraz je paradigma. V systéme Apertium, paradigma popisuje príklad, ako sa určitá skupina slov ohýba. V morfologickom slovníku sú lemmy spojené s paradigmami, ktoré nám dovolujú popisovať, ako sa daná lema ohýba bez toho, aby sme museli vypisovať všetky koncovky. The third word is paradigm. In the context of the Apertium system, paradigm refers to an example of how a particular group of words inflect. In the morphological dictionary, lemmas (see above) are linked to paradigms that allow us to describe how a given lemma inflects without having to write out all of the endings.

Ako príklad užitočnosti paradigiem, keby sme chceli uložiť dve anglické prídavné mená happy a lazy, namiesto okladania dva krát tej istej veci: An example of the utility of this is, if we wanted to store the two adjectives happy and lazy, instead of storing two lots of the same thing:

• happy, happ (y, ier, iest)
• lazy, laz (y, ier, iest)

Môžeme jednoducho uložiť jedno a potom povedať "lazy, skloňuje sa ako happy", ale aj shy inflects like happy", "naughty inflects like happy", "friendly inflects like happy", atď. V tomto príklade, happy by bolo paradigmou, modelom pre ohýbanie iných slov. Presná definícia bude vysvetlená o chvíľu. Paradigmy sú definovaný tagom <pardef>, používané tagom <par>. We can simply store one, and then say "lazy, inflects like happy", or indeed "shy inflects like happy", "naughty inflects like happy", "friendly inflects like happy", etc. In this example, happy would be the paradigm, the model for how the others inflect. The precise description of how this is defined will be explained shortly. Paradigms are defined in <pardef> tags, and used in <par> tags.

Jednojazyčné slovníky

Monolingual dictionaries

See also: List of dictionaries and Incubator

Začnime tvorbou nášho prvého slovníka. Slovník je XML súbor. Spustite váš textový editor a napíšte: Let's start by making our first source language dictionary. The dictionary is an XML file. Fire up your text editor and type the following:

<?xml version="1.0" encoding="UTF-8"?>
<dictionary>

</dictionary>

Zatiaľ tento súbor definuje, že chcemem vetvoriť slovník. Aby bol užitočný, treba pridať viac vstupov, prvým je abeceda. Definuje množinu písmen použivaných v Srbochorvátskom slovníku. Bude vyzerať nasledovne, obsahuje všetky znaky Srbochorvátskej abecedy: So, the file so far defines that we want to start a dictionary. In order for it to be useful, we need to add some more entries, the first is an alphabet. This defines the set of letters that may be used in the dictionary, for Serbo-Croatian. It will look something like the following, containing all the letters of the Serbo-Croatian alphabet:

<alphabet>ABCČĆDDžĐEFGHIJKLLjMNNjOPRSŠTUVZŽabcčćddžđefghijklljmnnjoprsštuvzž</alphabet>

Abecedu umiestnite pod tag <dictionary>. Place the alphabet below the <dictionary> tag.

Ďalej musíme definovať symboly. Začnime so základmi, podstatné meno (n) môže byť v singulári (sg) alebo v pluráli (pl). Next we need to define some symbols. Let's start off with the simple stuff, noun (n) in singular (sg) and plural (pl).

<sdefs>
   <sdef n="n"/>
   <sdef n="sg"/>
   <sdef n="pl"/>
</sdefs>

Mená symoblov nemusia byť krátke, no pretože ich budete opakovane písať, je výhodné ich skrátiť. The symbol names do not have to be so small, in fact, they could just be written out in full, but as you'll be typing them a lot, it makes sense to abbreviate.

Naneštastie, nie je to také jednoduché, podstatné mená v Srbochorváčtine skloňuje okrem čísla aj rod a pád. Pre náš príklad však budeme predpokladať, že podstatné mená budú mužského rodu v nominatíve (kompletný príklad nájdete na konci dokumentu). Unfortunately, it isn't quite so simple, nouns in Serbo-Croatian inflect for more than just number, they also inflect for gender and case. However, we'll assume for the purposes of this example that the noun is masculine and in the nominative case (a full example may be found at the end of this document).

Ďalej definujeme časť obsahujúcu paradigmy: The next thing is to define a section for the paradigms,

<pardefs>

</pardefs>

a slovníkovú časť: and a dictionary section:

<section id="main" type="standard">

</section>

Sú dva typy sekcií, štandardná sekcia obsahujúca slová, príklonky atď. Druhý typ je inconditional section, ktorá obyčajne obsahuje iterpunkciu atp. Takúto sekciu ešte nemáme. There are two types of sections, the first is a standard section, that contains words, enclitics, etc. The second type is an inconditional section which typically contains punctuation, and so forth. We don't have an inconditional section here.

Takže, náš súbor by mal vyzerať nasledovne: So, our file should now look something like:

<?xml version="1.0" encoding="UTF-8"?>
<dictionary>
   <sdefs>
     <sdef n="n"/>
     <sdef n="sg"/>
     <sdef n="pl"/>
   </sdefs>
   <pardefs>

   </pardefs>
   <section id="main" type="standard">

   </section>
</dictionary>

Keď teraz máme kostru, môžeme začať pridaním podstatného mena. Podstatné meno bude 'gramofon' (čo znamená gramofón). Now we've got the skeleton in place, we can start by adding a noun. The noun in question will be 'gramofon' (which means 'gramophone' or 'record player').

Prvú vec, ktorú musíme spraviť, je definovať paradigmy, pretožie ešte žiadne nemáme. The first thing we need to do, as we have no prior paradigms, is to define a paradigm.

Používame nominatív a mužský rod, pridáme teda len paradigmu pre singulár 'gramofon' a plurál 'gramofoni': Remember, we're assuming masculine gender and nominative case. The singular form of the noun is 'gramofon', and the plural is 'gramofoni'. So:

<pardef n="gramofon__n">
   <e><p><l/><r><s n="n"/><s n="sg"/></r></p></e>
   <e><p><l>i</l><r><s n="n"/><s n="pl"/></r></p></e>
</pardef>

Poznámka: tag '<l/>' (ekvivalnetné s <l></l>) znamená, že už netreba pridať ku kmeňu slova v singulári. Note: the '<l/>' (equivalent to <l></l>) denotes that there is no extra material to be added to the stem for the singular.

Možno to vyzerá ako celkom dlhý spôsob, ako to opísať, no je odôvodnený a rýchlo sa vám stane prirodzený. Zrejme vás zaujíma čo znamená <e>,

, <l> a <r>. Takže: This may seem like a rather verbose way of describing it, but there are reasons for this and it quickly becomes second nature. You're probably wondering what the <e>,

, <l> and <r> stand for. Well,

• e znamená záznam (entry).
• p znamená dvojicu (pair).
• l znamená ľavý (left).
• r znamená pravý (right).

• e, is for entry.
• p, is for pair.
• l, is for left.
• r, is for right.

Prečo pravý a ľavý? Morfologický slovík bude skompilovaný do konečného automatu. Kompilácia zľava doprava analyzuje slová, a kompilácia zprava doľava vytvára z analýzy slová. Napríklad: Why left and right? Well, the morphological dictionaries will later be compiled into finite state machines. Compiling them left to right produces analyses from words, and from right to left produces words from analyses. For example:

* gramofoni (left to right) gramofon<n><pl> (analýza)
* gramofon<n><pl> (right to left) gramofoni (tvorba)

Teraz sme definovali paradigmu, potrebujeme ju spojiť s príslošnou lemmou, gramofon. Vložíme to teda do sekcie ktorú sme definovali. Now we've defined a paradigm, we need to link it to its lemma, gramofon. We put this in the section that we've defined.

Záznam pre túto sekciu vyzerá nasledovne: The entry to put in the </dictionary>

Teraz potrebujeme pridať záznam na preklad medzi dvoma slovami: Now we need to add an entry to translate between the two words. Something like:

<e><p><l>gramofon<s n="n"/></l><r>gramophone<s n="n"/></r></p></e>

Pretože tu bude veľa takýchto záznamov, bývajú obyčajne kvôli čitateľnosti písané na jeden riadok. Znova používame 'l' a 'r'. Pri kompilácii zľava dostaneme preklad Srbochorváčtina → Angličtina a pri kompilácii zprava dostaneme preklad Angličtina → Srbochorváčtina. Because there are a lot of these entries, they're typically written on one line to facilitate easier reading of the file. Again with the 'l' and 'r' right? Well, we compile it left to right to produce the Serbo-Croatian → English dictionary, and right to left to produce the English → Serbo-Croatian dictionary.

Keď je toto hotové, spustite nasledujúce príkazy: So, once this is done, run the following commands:

$ lt-comp lr apertium-sh-en.sh.dix sh-en.automorf.bin
$ lt-comp rl apertium-sh-en.en.dix sh-en.autogen.bin

$ lt-comp lr apertium-sh-en.en.dix en-sh.automorf.bin
$ lt-comp rl apertium-sh-en.sh.dix en-sh.autogen.bin

$ lt-comp lr apertium-sh-en.sh-en.dix sh-en.autobil.bin
$ lt-comp rl apertium-sh-en.sh-en.dix en-sh.autobil.bin

Aby ste vygenerovali morfologické analyzátory (automorf), morfologické generátory (autogen) a vyhľadávač slov (autobil, bil ako bilingválny). To generate the morphological analysers (automorf), the morphological generators (autogen) and the word lookups (autobil), the bil is for "bilingual".

Transfer rules

Teraz máme dva morfologické slovníka a dvojjazyčný slovník. Všetko, čo potrebujeme, sú pravidlá na prenos podstatných mien. Súbory popisujúce tieto pravidlá majú vlstný DTD (transfer.dtd), ktorý môžete nájsť v balíčku Apertium. Ak potrebujete vytvoriť pravidlo, často je dobrý nápad nazrieť do súborov pravidiel iných jazykových dvojíc. Veľa pravidiel môže byť znovu použitých aj medzi jazykmi. Napríklad, tento konkrétky popísaný nižie by bol užitočný pre každý jazyk so zamlčaným podmetom So, now we have two morphological dictionaries, and a bilingual dictionary. All that we need now is a transfer rule for nouns. Transfer rule files have their own DTD (transfer.dtd) which can be found in the Apertium package. If you need to implement a rule it is often a good idea to look in the rule files of other language pairs first. Many rules can be recycled/reused between languages. For example the one described below would be useful for any null-subject language.

Začnime ako obyčajne so základnou kostrou: Start out like all the others with a basic skeleton:

<?xml version="1.0" encoding="UTF-8"?>
<transfer>

</transfer>

V tejto chvíli, pretože ignorujeme pády, stačí len vytvoriť pravidlo, ktoré načíta gramatické symboly a zase ich vypíše. At the moment, because we're ignoring case, we just need to make a rule that takes the grammatical symbols input and outputs them again.

Najprv musíme definovať kategórie a atribúty. Obe nám dovoľujú spájať gramatické symboly. Kategórie umožňujú spájať symboly za účelom párovania (napríklad 'n.*' sú všetky podstatné mená). Atribúty zase spájajú skupiny symbolov, z ktorých sa môžu symboly vyberať, napríklad 'sg' a 'pl' môžu byť spojené v atribúte číslo ('number'). We first need to define categories and attributes. Categories and attributes both allow us to group grammatical symbols. Categories allow us to group symbols for the purposes of matching (for example 'n.*' is all nouns). Attributes allow us to group a set of symbols that can be chosen from. For example ('sg' and 'pl' may be grouped a an attribute 'number').

Pridajme teda potrebné sekcie: Lets add the necessary sections:

<section-def-cats>

</section-def-cats>
<section-def-attrs>

</section-def-attrs>

Pretože iba skloňujeme podstatné mená v singulári a pluráli, potrebujeme pridať kategóriu pre podstatné mena a atribút pre číslo: As we're only inflecting, nouns in singular and plural then we need to add a category for nouns, and with an attribute of number. Something like the following will suffice:

Do section-def-cats pridajte: Into section-def-cats add:

<def-cat n="nom">
   <cat-item tags="n.*"/>
</def-cat>

Toto zachytí všetky podstatné mená (lemmy nasledované <n>, potom hocičím) a pomenuje ich "nom" (neskôr to využijeme). This catches all nouns (lemmas followed by <n> then anything) and refers to them as "nom" (we'll see how thats used later).

Do sekcie section-def-attrs, pridajte: Into the section section-def-attrs, add:

<def-attr n="nbr">
   <attr-item tags="sg"/>
   <attr-item tags="pl"/>
</def-attr>

a potom and then

<def-attr n="a_nom">
   <attr-item tags="n"/>
</def-attr>

Prvý definuje atribút nbr (číslo), ktoré môže byť buď singulár alebo plurál. The first defines the attribute nbr (number), which can be either singular (sg) or plural (pl).

Druhý definuje atribút a_nom (atribút podstatné meno). The second defines the attribute a_nom (attribute noun).

Ďalej potrebujeme pridať sekciu pre globálne premenné: Next we need to add a section for global variables:

<section-def-vars>

</section-def-vars>

Tieto premenné sú používané na ukladanie alebo prenos atribútov medzi pravidlami. Na teraz potrebujeme len jedno, These variables are used to store or transfer attributes between rules. We need only one for now,

<def-var n="number"/>

Nakoniec, potrebujeme pridať pravidlo, ktoré vyberie podstatné meno a vypíše ho v správnom tvare. Budeme potrebovať sekciu pravidiel... Finally, we need to add a rule, to take in the noun and then output it in the correct form. We'll need a rules section...

<section-rules>

</section-rules>

Zmením tempo z predchádzajúcich príkladov, len sem vložím toto pravidlo a potom ho opíšem. Changing the pace from the previous examples, I'll just paste this rule, then go through it, rather than the other way round.

<rule>
   <pattern>
     <pattern-item n="nom"/>
   </pattern>
   <action>
     <out>
       <lu>
         <clip pos="1" side="tl" part="lem"/>
         <clip pos="1" side="tl" part="a_nom"/>
         <clip pos="1" side="tl" part="nbr"/>
       </lu>
     </out>
   </action>
</rule>

Prvý tag je zrejmý, definuje pravidlo. Druhý tag, pattern, v zásade hovorí: "použi toto pravidlo, ak nájdeš tento vzor". V tomto príklade tento vzor pozostáva z jediného podstatného mena (definovného kategóriou nom). Všimnite si, že vzor je napárovaný s najdlším výskytom najprv. Takže ak máte tri pravidlá, prvé zachytáva "<prn><vblex><n>", druhé zachytáva "<prn><vblex>" tretie zachytáva "<n>", a bol zachytený vzor, a vykonané pravidlo bude to prvé. The first tag is obvious, it defines a rule. The second tag, pattern basically says: "apply this rule, if this pattern is found". In this example the pattern consists of a single noun (defined by the category item nom). Note that patterns are matched in a longest-match first. So if you have three rules, the first catches "<prn><vblex><n>", the second catches "<prn><vblex>" and the third catches "<n>", the pattern matched, and rule executed will be the first.

Pre každý vzor existuje akcia, ktorá produkuje výstup. Tento výstup je lexikálna jednotka (lu). For each pattern, there is an associated action, which produces an associated output, out. The output, is a lexical unit (lu).

Tag clip umožnuje používateľovi vybrať a manipulovať atribúty a časti zdrojového jazyka (side="sl"), alebo lexikálne poločky cielového jazyka (side="tl"). The clip tag allows a user to select and manipulate attributes and parts of the source language (side="sl"), or target language (side="tl") lexical item.

Skompilujme a otestuje to. Pravidlá prenosu sú skompilované s : Let's compile it and test it. Transfer rules are compiled with:

$ apertium-preprocess-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin

Čo vygeneruje súbor sh-en.t1x.bin. Which will generate a sh-en.t1x.bin file.

Teraz sme takmer pripravení otestovať náš systém pre strojový preklad. Chýba jedna dôležitá časť na určenie slovných druhov (tzv. PoS tagger), no to bude rýchlo vysvetlené. Zatiaľ ho môžeme otestovať v takomto stave: Now we're ready to test our machine translation system. There is one crucial part missing, the part-of-speech (PoS) tagger, but that will be explained shortly. In the meantime we can test it as is:

Najprv, analyzujme slovo, gramofoni: First, lets analyse a word, gramofoni:

$ echo "gramofoni" | lt-proc sh-en.automorf.bin 
^gramofon/gramofon<n><pl>$

Teraz by PoS tagger vybral správnu verziu podľa slovného druhu, ale keďže ho ešte nemáme, použijeme malý gawk script (vďaka Sergiovi) ktorý vypíše prvú možnosť. Now, normally here the POS tagger would choose the right version based on the part of speech, but we don't have a POS tagger yet, so we can use this little gawk script (thanks to Sergio) that will just output the first item retrieved.

$ echo "gramofoni" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}'
^gramofon<n><pl>$

Teraz pokračujme s pravidlom pre prenos: Now let's process that with the transfer rule:

$ echo "gramofoni" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin

Čo vypíše: It will output:

^gramophone<n><pl>$^@

• 'gramophone' anguage (side="tl") lemma (lem) at position 1 (pos="1").
• '<n>' is the target language a_nom at position 1.
• '<pl>' is the target language attribute of number (nbr) at position 1.

• 'gramophone' is the target language (side="tl") lemma (lem) at position 1 (pos="1").
• '<n>' is the target language a_nom at position 1.
• '<pl>' is the target language attribute of number (nbr) at position 1.

Try commenting out one of these clip statements, recompiling and seeing what happens.

So, now we have the output from the transfer, the only thing that remains is to generate the target-language inflected forms. For this, we use lt-proc, but in generation (-g), not analysis mode.

$ echo "gramofoni" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin | \
  lt-proc -g sh-en.autogen.bin

gramophones\@

And c'est ca. You now have a machine translation system that translates a Serbo-Croatian noun into an English noun. Obviously this isn't very useful, but we'll get onto the more complex stuff soon. Oh, and don't worry about the '@' symbol, I'll explain that soon too.

Think of a few other words that inflect the same as gramofon. How about adding those. We don't need to add any paradigms, just the entries in the main section of the monolingual and bilingual dictionaries.

Bring on the verbs

Ok, so we have a system that translates nouns, but thats pretty useless, we want to translate verbs too, and even whole sentences! How about we start with the verb to see. In Serbo-Croatian this is videti. Serbo-Croatian is a null-subject language, this means that it doesn't typically use personal pronouns before the conjugated form of the verb. English is not. So for example: I see in English would be translated as vidim in Serbo-Croatian.

• Vidim
• see<p1><sg>
• I see

Note: <p1> denotes first person

This will be important when we come to write the transfer rule for verbs. Other examples of null-subject languages include: Spanish, Romanian and Polish. This also has the effect that while we only need to add the verb in the Serbo-Croatian morphological dictionary, we need to add both the verb, and the personal pronouns in the English morphological dictionary. We'll go through both of these.

The other forms of the verb videti are: vidiš, vidi, vidimo, vidite, and vide; which correspond to: you see (singular), he sees, we see, you see (plural), and they see.

There are two forms of you see, one is plural and formal singular (vidite) and the other is singular and informal (vidiš).

We're going to try and translate the sentence: "Vidim gramofoni" into "I see gramophones". In the interests of space, we'll just add enough information to do the translation and will leave filling out the paradigms (adding the other conjugations of the verb) as an exercise to the reader.

The astute reader will have realised by this point that we can't just translate vidim gramofoni because it is not a grammatically correct sentence in Serbo-Croatian. The correct sentence would be vidim gramofone, as the noun takes the accusative case. We'll have to add that form too, no need to add the case information for now though, we just add it as another option for plural. So, just copy the 'e' block for 'i' and change the 'i' to 'e' there.

First thing we need to do is add some more symbols. We need to first add a symbol for 'verb', which we'll call "vblex" (this means lexical verb, as opposed to modal verbs and other types). Verbs have 'person', and 'tense' along with number, so lets add a couple of those as well. We need to translate "I see", so for person we should add "p1", or 'first person', and for tense "pri", or 'present indicative'.

<sdef n="vblex"/>
<sdef n="p1"/>
<sdef n="pri"/>

After we've done this, the same with the nouns, we add a paradigm for the verb conjugation. The first line will be:

<pardef n="vid/eti__vblex">

The '/' is used to demarcate where the stems (the parts between the <l> </l> tags) are added to.

Then the inflection for first person singular:

<e><p><l>im</l><r>eti<s n="vblex"/><s n="pri"/><s n="p1"/><s n="sg"/></r></p></e>

The 'im' denotes the ending (as in 'vidim'), it is necessary to add 'eti' to the <r> section, as this will be chopped off by the definition. The rest is fairly straightforward, 'vblex' is lexical verb, 'pri' is present indicative tense, 'p1' is first person and 'sg' is singular. We can also add the plural which will be the same, except 'imo' instead of 'im' and 'pl' instead of 'sg'.

After this we need to add a lemma, paradigm mapping to the main section:

<e lm="videti"><i>vid</i><par n="vid/eti__vblex"/></e>

Note: the content of <i> </i> is the root, not the lemma.

Thats the work on the Serbo-Croatian dictionary done for now. Lets compile it then test it.

$ lt-comp lr apertium-sh-en.sh.dix sh-en.automorf.bin
main@standard 23 25
$ echo "vidim" | lt-proc sh-en.automorf.bin
^vidim/videti<vblex><pri><p1><sg>$
$ echo "vidimo" | lt-proc sh-en.automorf.bin
^vidimo/videti<vblex><pri><p1><pl>$

Ok, so now we do the same for the English dictionary (remember to add the same symbol definitions here as you added to the Serbo-Croatian one).

The paradigm is:

<pardef n="s/ee__vblex">

because the past tense is 'saw'. Now, we can do one of two things, we can add both first and second person, but they are the same form. In fact, all forms (except third person singular) of the verb 'to see' are 'see'. So instead we make one entry for 'see' and give it only the 'pri' symbol.

<e><p><l>ee</l><r>ee<s n="vblex"/><s n="pri"/></r></p></e>

and as always, an entry in the main section:

<e lm="see"><i>s</i><par n="s/ee__vblex"/></e>

Then lets save, recompile and test:

$ lt-comp lr apertium-sh-en.en.dix en-sh.automorf.bin
main@standard 18 19

$ echo "see" | lt-proc en-sh.automorf.bin
^see/see<vblex><pri>$

Now for the obligatory entry in the bilingual dictionary:

<e><p><l>videti<s n="vblex"/></l><r>see<s n="vblex"/></r></p></e>

(again, don't forget to add the sdefs from earlier)

And recompile:

$ lt-comp lr apertium-sh-en.sh-en.dix sh-en.autobil.bin
main@standard 18 18
$ lt-comp rl apertium-sh-en.sh-en.dix en-sh.autobil.bin
main@standard 18 18

Now to test:

$ echo "vidim" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin

^see<vblex><pri><p1><sg>$^@

We get the analysis passed through correctly, but when we try and generate a surface form from this, we get a '#', like below:

$ echo "vidim" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin | \
  lt-proc -g sh-en.autogen.bin
#see\@

This '#' means that the generator cannot generate the correct lexical form because it does not contain it. Why is this?

Basically the analyses don't match, the 'see' in the dictionary is see<vblex><pri>, but the see delivered by the transfer is see<vblex><pri><p1><sg>. The Serbo-Croatian side has more information than the English side requires. You can test this by adding the missing symbols to the English dictionary, and then recompiling, and testing again.

However, a more paradigmatic way of taking care of this is by writing a rule. So, we open up the rules file (apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin in case you forgot).

We need to add a new category for 'verb'.

<def-cat n="vrb">
   <cat-item tags="vblex.*"/>
</def-cat>

We also need to add attributes for tense and for person. We'll make it really simple for now, you can add p2 and p3, but I won't in order to save space.

<def-attr n="temps">
   <attr-item tags="pri"/>
</def-attr>

<def-attr n="pers">
   <attr-item tags="p1"/>
</def-attr>

We should also add an attribute for verbs.

<def-attr n="a_verb">
   <attr-item tags="vblex"/>
</def-attr>

Now onto the rule:

<rule>
   <pattern>
     <pattern-item n="vrb"/>
   </pattern>
   <action>
     <out>
       <lu>
         <clip pos="1" side="tl" part="lem"/>
         <clip pos="1" side="tl" part="a_verb"/>
         <clip pos="1" side="tl" part="temps"/>
       </lu>
     </out>
   </action>
</rule>

Remember when you tried commenting out the 'clip' tags in the previous rule example and they disappeared from the transfer, well, thats pretty much what we're doing here. We take in a verb with a full analysis, but only output a partial analysis (lemma + verb tag + tense tag).

So now, if we recompile that, we get:

$ echo "vidim" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin
^see<vblex><pri>$^@

and:

$ echo "vidim" | lt-proc sh-en.automorf.bin  | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin | \
  lt-proc -g sh-en.autogen.bin
see\@

Try it with 'vidimo' (we see) to see if you get the correct output.

Now try it with "vidim gramofone":

$ echo "vidim gramofoni" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin | \
  lt-proc -g sh-en.autogen.bin
see gramophones\@

But what about personal pronouns?

Well, thats great, but we're still missing the personal pronoun that is necessary in English. In order to add it in, we first need to edit the English morphological dictionary.

As before, the first thing to do is add the necessary symbols:

<sdef n="prn"/>
<sdef n="subj"/>

Of the two symbols, prn is pronoun, and subj is subject (as in the subject of a sentence).

Because there is no root, or 'lemma' for personal subject pronouns, we just add the pardef as follows:

<pardef n="prsubj__prn">
   <e><p><l>I</l><r>prpers<s n="prn"/><s n="subj"/><s n="p1"/><s n="sg"/></r></p></e>
</pardef>

With 'prsubj' being 'personal subject'. The rest of them (You, We etc.) are left as an exercise to the reader.

We can add an entry to the main section as follows:

<e lm="personal subject pronouns"><i/><par n="prsubj__prn"/></e>

So, save, recompile and test, and we should get something like:

$ echo "I" | lt-proc en-sh.automorf.bin
^I/PRPERS<prn><subj><p1><sg>$

(Note: it's in capitals because 'I' is in capitals).

Now we need to amend the 'verb' rule to output the subject personal pronoun along with the correct verb form.

First, add a category (this must be getting pretty pedestrian by now):

<def-cat n="prpers">
   <cat-item lemma="prpers" tags="prn.*"/>
</def-cat>

Now add the types of pronoun as attributes, we might as well add the 'obj' type as we're at it, although we won't need to use it for now:

<def-attr n="tipus_prn">
   <attr-item tags="prn.subj"/>
   <attr-item tags="prn.obj"/>
</def-attr>

And now to input the rule:

<rule>
   <pattern>
     <pattern-item n="vrb"/>
   </pattern>
   <action>
     <out>
       <lu>
         <lit v="prpers"/>
         <lit-tag v="prn"/>
         <lit-tag v="subj"/>
         <clip pos="1" side="tl" part="pers"/>
         <clip pos="1" side="tl" part="nbr"/>
       </lu>
       <b/>
       <lu>
         <clip pos="1" side="tl" part="lem"/>
         <clip pos="1" side="tl" part="a_verb"/>
         <clip pos="1" side="tl" part="temps"/>
       </lu>
     </out>
   </action>
</rule>

This is pretty much the same rule as before, only we made a couple of small changes.

We needed to output:

^prpers<prn><subj><p1><sg>$ ^see<vblex><pri>$

so that the generator could choose the right pronoun and the right form of the verb.

So, a quick rundown:

• <lit>, prints a literal string, in this case "prpers"
• <lit-tag>, prints a literal tag, because we can't get the tags from the verb, we add these ourself, "prn" for pronoun, and "subj" for subject.
• , prints a blank, a space.

Note that we retrieved the information for number and tense directly from the verb.

So, now if we recompile and test that again:

$ echo "vidim gramofone" | lt-proc sh-en.automorf.bin  | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin | \
  lt-proc -g sh-en.autogen.bin
I see gramophones

Which, while it isn't exactly prize-winning prose (much like this HOWTO), is a fairly accurate translation.

So tell me about the record player (Multiwords)

While gramophone is an English word, it isn't the best translation. Gramophone is typically used for the very old kind, you know with the needle instead of the stylus, and no powered amplification. A better translation would be 'record player'. Although this is more than one word, we can treat it as if it is one word by using multiword (multipalabra) constructions.

We don't need to touch the Serbo-Croatian dictionary, just the English one and the bilingual one, so open it up.

The plural of 'record player' is 'record players', so it takes the same paradigm as gramophone (gramophone__n) — in that we just add 's'. All we need to do is add a new element to the main section.

<e lm="record player"><i>record<b/>player</i><par n="gramophone__n"/></e>

The only thing different about this is the use of the tag, although this isn't entirely new as we saw it in use in the rules file.

So, recompile and test in the orthodox fashion:

$ echo "vidim gramofone" | lt-proc sh-en.automorf.bin | \
  gawk 'BEGIN{RS="$"; FS="/";}{nf=split($1,COMPONENTS,"^"); for(i = 1; i<nf; i++) printf COMPONENTS[i]; if($2 != "") printf("^%s$",$2);}' | \
  apertium-transfer apertium-sh-en.sh-en.t1x sh-en.t1x.bin sh-en.autobil.bin  | \
  lt-proc -g sh-en.autogen.bin
I see record players

Perfect. A big benefit of using multiwords is that you can translate idiomatic expressions verbatim, without having to do word-by-word translation. For example the English phrase, "at the moment" would be translated into Serbo-Croatian as "trenutno" (trenutak = moment, trenutno being adverb of that) — it would not be possible to translate this English phrase word-by-word into Serbo-Croatian.

Dealing with minor variation

Serbo-Croatian typically has a few ways of writing each word because of dialectal variation. It has a cool phonetic writing system so you write how you speak. For example, people speaking in Ijekavian would say "rječnik", while someone speaking Ekavian would say "rečnik", which reflects the differences in pronunciation of the proto-Slavic vowel yat.

Analysis

There should be a fairly easy way of dealing with this, and there is, using paradigms again. Paradigms aren't only used for adding grammatical symbols, but they can also be used to replace any character/symbol with another. For example, here is a paradigm for accepting both "e" and "je" in the analysis. The paradigm should, as with the others go into the monolingual dictionary for Serbo-Croatian.

  <pardef n="e_je__yat">
    <e>
      <p>
        <l>e</l>
        <r>e</r>
      </p>
    </e>
    <e>
      <p>
        <l>je</l>
        <r>e</r>
      </p>
    </e>
  </pardef>

Then in the "main section":

    <e lm="rečnik"><i>r</i><par n="e_je__yat"/><i>čni</i><par n="rečni/k__n"/></e>

This only allows us to analyse both forms however... more work is necessary if we want to generate both forms.

See also

• Building dictionaries
• Cookbook
• Chunking
• Contributing to an existing pair