Working With Glossaries

Glossaries

A glossary is a file in plain text (UTF-8 encoded) containing bilingual alignments to force specific translations. When a glossary is specified and the text to translate includes one or more phrases from the first part of the glossary file, the translation will be forced to the second part.

Example for a Spanish to English glossary:

Pedro       Pedro

Juan        Juan

Important: The first and second phrases in each line must be separated by a TAB character.

This means that any occurrences of "Pedro" and "Juan" will remain unchanged during the translation, instead of being translated to "Peter" and "John".

Glossaries are typically used to preserve proper names, brand names, and people’s names. They are not a replacement for Translation Memories and have limited applicability in inflected languages.

Creating a Glossary

Glossary files are created by linguists using plain text editors such as Sublime Text or Notepad++.

Key considerations:

• Always use UTF-8 encoding, especially when working with special characters.

• Use a TAB character to separate the two terms on each line.

• Glossary entries are case-sensitive. If needed, use different lines for different casing.

• Save the file with a recognizable name and a .txt extension (e.g., my-first-glossary.txt).

Managing Glossaries in ECO

Once the glossary is ready as a .txt file, you must upload it to ECO using an admin account. Regular users can use glossaries but cannot manage them.

Navigate to the left menu:
Corporate > Dicts/Glossaries

In this section, you can upload, delete, and assign glossaries to engines.

Note: Glossaries and Dictionaries can be used for more than translation. Be sure to use the type: GLOSSARY / Translation.

To use a glossary during translation, it must be assigned to one or more engines.

Managing Glossaries in ECO

Once the glossary is ready as a .txt file, you must upload it to ECO using an admin account. Regular users can use glossaries but cannot manage them.

Navigate to the left menu:
Corporate > Dicts/Glossaries

In this section, you can upload, delete, and assign glossaries to engines.

Note: Glossaries and Dictionaries can be used for more than translation. Be sure to use the type: GLOSSARY / Translation.

To use a glossary during translation, it must be assigned to one or more engines.

Using a Glossary When Processing a File in ECO

You can specify a glossary when translating an array of segments via API by including its unique identifier (glossary_id, an integer).

To get the list of available glossaries for your API key, use the following endpoint:

curl --location --request POST 'http://prod.pangeamt.com:8080/NexRelay/v1/corp/glossaries' \

--header 'Content-Type: application/json' \

--data-raw '{"apikey":"demo"}'

You must include your APIKey in the request.

Response Example

[

  {

    "id": 3,

    "name": "Automotive",

    "engineids": [15]

  },

  {

    "id": 4,

    "name": "Medical_1",

    "engineids": [12, 10]

  },

  {

    "id": 5,

    "name": "Medical_2",

    "engineids": [12, 10, 934]

  }

]

Each glossary is identified by its id and the list of engines it's assigned to.

Example: Translating Using a Glossary

If you want to translate with engine 12 and glossary “Medical_2” (ID 5), use the following request body when calling the translate API:

{

  "src": "es",

  "tgt": "en",

  "apikey": "demo",

  "engine": "12",

  "text": ["Casos de uso", "Esto es hola mundo!"],

  "glossary_id": 5

}