Aligner

Aligner

The Aligner is a tool supports the integration of texts into the translation memory. Strictly speaking it supports the alignment, that is the separation of translation examples from texts that are available in two languages. It thus enables the user to handle pairs of bilingual texts.

Bilingual texts are a valuable source for translators. However, the sentences in the texts need to be separated to serve as translation examples. More important, separation must reflect the corresponding meaning in the separated sentences. Not every single sentence is translated into another language by exactly another single sentence. Sometimes one sentence is translated by two sentences or even three. It’s hard to completely automate this separation process. Without user control a correct alignment is barely possible.

Aligning the bilingual texts
Aligner allows the user not only to separate sentences but also to correct typos, obvious misspellings, wrong hyphenations and the like.
Aligner separates the corresponding sentences along the dot and shows the translations in pairs on the left side one below the other. The following translation pair is shown on the right side. With this interface the user can check and correct the separation of sentence pairs.

separate button | join button

• Setting a separation mark (marking any spot in the sentence) on one of the text fields on the left and clicking the separate button sends the sentence fragment after the click position to the right hand side.
• Clicking the join button sends the sentence (fragment) from the right hand side to left and adds it to the sentence shown there.

In the example above, the sentence in the right lower field must be translated by the two sentence fragments in the two upper text fields. By clicking on the join button the sentence (fragment) on the right hand side is added to the sentence (fragment) on the left side.

To undo the join/separation of sentences or sentence fragments the user just presses the undo button.

undo Button

The user can also mark a text section, copy or cut it, and finally paste it in the opposite text field.

After marking a text section, the two options copy and cut become available:
After copying or cutting a text section, the paste operation is enabled in the opposite text field:

The user navigates through the text using the left and right arrows. Any modifications on the text sections on both sides are automatically saved when navigating to the next/previous sentence:

Setting the workspace

To start the alignment the user has to set the workspace first. Aligner remembers the settings and sets the actual workspace as default workspace in subsequent sessions.
To switch the workspace the user opens the directory containing the text documents that need to be integrated. The name of the workspace (directory) is indicated right of the button.

Selecting the texts

Pressing the button with the list opens a dialogue showing the list of (bilingual/multilingual) text documents that can be aligned and integrated into the translation memory.
The left button opens the dialog to select the texts to be processed. The right one adds the newly processed entries to the translation memory.

The dialogue shows the titles of the text documents that are candidates for the text integration i.e. that are available in two or more languages.
Each list entry shows the title in the different languages, the available language, the unique identifier of the text and the date when the text has been processed.

The indication of the date is important to inform the user about a text that has been aligned in the past. The translation memory already contains its sentences.

Important: The text documents must be available in UTF-8 encoding with file name extensions “txt” or “text”.
Furthermore, it is important to indicate the available language in the texts. There are two ways:
The file name gets the two-letter language identifier before the file extension, like “document.en.txt” or “document.de.text”.
The first line in the text contains the locale identifier: “en-US” or “de-DE”.

The unique identifier of the text is the file name without language identifier and file name extension, like “document” in the example above.

Installation Guide
Copy Aligner.jar to the location of your choice.
In the same directory you will find a text file called aligner.config when you run the Aligner for the first time. You cannot edit its content and usually there is no need to do so.

Requirements
Java installed on your machine.