Lilt API API Reference
The Lilt API enables programmatic access to the full-range of Lilt backend services including:
- Training of and translating with interactive, adaptive machine translation
- Large-scale translation memory
- The Lexicon (a large-scale termbase)
- Programmatic control of the Lilt CAT environment
- Translation memory synchronization
Requests and responses are in JSON format. The API only responds to HTTPS / SSL requests. The API authenticates via an API key, which is included with the Business plan.
API Endpoint
https://lilt.com/2Contact: support@lilt.com
Schemes: https
Version: v2.0
Authentication
api_key
Root
Retrieve the API root
This resource does not have any attributes. It lists the name of the API. This endpoint can be used to verify API keys and to check the availability of the API.
OK
Documents
Create a Document
Create a new Document. A Document is a collection of one or more Segments.
Documents are nested inside of Projects, and appear in the Project details view in the web app.
Document-level relationships between Segments are considered by the machine translation system during adaptation. If there is no inherent document structure in your data, you still might consider grouping related Segments into Documents to improve translation quality.
The Document resource to create.
- name: string
-
The document name.
- project_id: integer
-
A unique Project identifier.
Request Example
{
"name": "Introduction.xliff",
"project_id": 23618
}
OK
Response Example (200 OK)
{
"id": 46530,
"srclang": "en",
"trglang": "de",
"name": "Introduction.xliff",
"segments": [
"integer"
]
}
Retrieve a Document
List a Document.
The unique Document identifier.
OK
Response Example (200 OK)
{
"id": 46530,
"srclang": "en",
"trglang": "de",
"name": "Introduction.xliff",
"segments": [
"integer"
]
}
Update a Document
Update a Document.
The Document resource to update.
- id: integer
-
A unique Document identifier.
- name: string
-
The Document name.
Request Example
{
"id": 46530,
"name": "Introduction to our App"
}
OK
Response Example (200 OK)
{
"id": 46530,
"name": "Introduction to our App"
}
Delete a Document
Delete a Document.
The unique Document identifier.
OK
Assign a Document
Assign and unassign a Document for translation and/or review.
Attributes of the Document resource to assign.
- id: integer
-
A unique Document identifier.
- email: string
-
An email address.
- is_translator: boolean
-
If true, assign for translating. If false, then unassign.
- is_reviewer: boolean
-
If true, assign for reviewing. If false, then unassign.
Request Example
{
"id": 46530,
"email": "user@email.com",
"is_translator": true,
"is_reviewer": "boolean"
}
OK
Response Example (200 OK)
{
"id": 46530
}
Download a File
Download a Document in XLIFF 1.2 format.
Example CURL command:
curl -X GET https://lilt.com/2/documents/files?key=API_KEY&id=274 -o from_lilt.xliff
An unique Document identifier.
The response is an XLIFF 1.2 file.
Response Example (200 OK)
"my_document.xliff"
Upload a File
Create a Document from an XLIFF 1.2 file. Header parameters should be passed as JSON object with the header field LILT-API.
Example CURL command:
curl -X POST https://lilt.com/2/documents/files?key=API_KEY \
--header "LILT-API: {\"name\": \"introduction.xliff\",\"project_id\": 9}" \
--header "Content-Type: application/octet-stream" \
--data-binary @Introduction.xliff
The file contents to be uploaded. The entire POST body will be treated as the file.
A file name.
A project id.
Request Example
"<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n<xliff xmlns=\"urn:oasis:names:tc:xliff:document:1.2\"> \n...\n</xliff>"
OK
Response Example (200 OK)
{
"id": 46530,
"srclang": "en",
"trglang": "de",
"name": "Introduction.xliff",
"segments": [
"integer"
]
}
Languages
Retrieve supported languages
Get a list of supported languages.
Locales (e.g., en-US) are not explicitly supported. To create locale-specific systems, simply create a new Memory and upload locale-specific translation memory data to it. The system will adapt to the locale.
OK
Response Example (200 OK)
{
"source_to_target": {
"en": {
"da": true,
"de": true,
"fr": true,
"...": "..."
},
"...": "..."
},
"code_to_name": {
"aa": "Afar",
"ab": "Abkhazian",
"af": "Afrikaans",
"...": "..."
}
}
Lexicon
Update a Lexicon
Update the Lexicon with a new term. The maximum source length is 250 characters.
The lexicon entry.
- memory_id: integer
-
A unique Memory identifier.
- source: string
-
The source side of the lexicon entry.
- target: string
-
The target side of the lexicon entry.
Request Example
{
"memory_id": 10641,
"source": "washing machine",
"target": "Waschmaschine"
}
OK
Query a Lexicon
Query the Lexicon. The Lexicon is an editable termbase / concordance that is integrated with the Memory.
A unique Memory identifier.
An ISO639-1 language code.
An ISO639-1 language code.
The query term.
The maximum number of results to return.
OK
Response Example (200 OK)
[
{
"translations": [],
"examples": [
[
{
"source": "Waschmaschine",
"sourceDelimiters": [
"",
""
],
"target": [
"washing",
"machine"
],
"targetDelimiters": [
"",
" ",
""
],
"sourceSpan": {
"start": 0,
"length": 1
},
"targetSpan": {
"start": 1,
"length": 1
},
"similarity": 0,
"memoryId": 10641
}
]
]
}
]
Memories
Create a Memory
Create a new Memory. A Memory is a container that collects source/target sentences for a specific language pair (e.g., English>French). The data in the Memory is used to train the MT system, populate the TM, and update the lexicon. Memories are private to your account - the data is not shared across users - unless you explicitly share a Memory with your team (via web app only). Please refer to our KB for a more detailed description.
The Memory resource to create.
- name: string
-
A name for the Memory.
- srclang: string
-
An ISO639-1 language identifier.
- trglang: string
-
An ISO639-1 language identifier.
Request Example
{
"name": "automotive",
"srclang": "en",
"trglang": "fr"
}
OK
Response Example (200 OK)
{
"id": 7245,
"name": "automotive",
"num_segments": 890,
"srclang": "en",
"trglang": "fr"
}
Retrieve a Memory
Retrieve a Memory. If you cannot access the Memory (401 error) please check permissions (e.g. in case you created the Memory via the web app with a different account you may have to explicitly share that Memory).
An optional Memory identifier.
OK
Response Example (200 OK)
[
{
"id": 7245,
"name": "automotive",
"num_segments": 890,
"srclang": "en",
"trglang": "fr"
}
]
Update a Memory
Update a Memory.
The Memory resource to update.
- id: integer
-
A unique Memory identifier.
- name: string
-
The Memory name.
Request Example
{
"id": 7246,
"name": "Automotive Memory"
}
OK
Response Example (200 OK)
{
"id": 7246,
"name": "Automotive Memory"
}
Delete a Memory
Delete a Memory.
A unique Memory identifier.
OK
Query a Memory
Perform a translation memory query.
A unique Memory identifier.
A source query.
Maximum number of results to return.
OK
Response Example (200 OK)
[
{
"created_at": "string (date)",
"memory_id": "integer",
"source": "string",
"created_by": {
"email": "string",
"name": "string"
}
}
]
Get-sync for a Memory
Get all or part of a memory in TMX 1.4b format. If the from_time and/or to_time are omitted, then all segments are returned. The parameter when specifies on which date field from_time and to_time are matched. Possible values are Created (when the segment was originally created in the memory), Updated (when the segment was lastly updated), and Deleted (when the segment was deleted).
Example CURL command:
curl -X GET https://lilt.com/2/memories/sync?key=API_KEY&id=42 -o from_lilt.tmx
A unique Memory identifier.
Unix time stamp (epoch, in seconds) of the start of the Memory section.
Unix time stamp (epoch, in seconds) of the end of the Memory section.
The date field on which retrieved segments match from/to time stamps: Created, Updated, Deleted. If this parameter is omitted, then the whole Memory is returned.
The response is a TMX 1.4b file.
Response Example (200 OK)
"string"
Insert-sync for a Memory
Inserts a TM in TMX 1.4b format into the Memory. Header parameters should be passed as JSON object with the header field LILT-API.
Example CURL command:
curl -X POST https://lilt.com/2/memories/sync?key=API_KEY \
--header "LILT-API: {\"name\": \"my_memory.tmx\",\"id\": 42}" \
--header "Content-Type: application/octet-stream" \
--data-binary @my_memory.tmx
The file contents to be uploaded. The entire POST body will be treated as the file.
A unique Memory identifier.
Name of the TMX file.
Request Example
"<?xml version=\"1.0\" encoding=\"UTF-8\" ?>\n<tmx version=\"1.4\">\n...\n</tmx>"
OK
Response Example (200 OK)
{
"id": "integer",
"num_updates": "integer"
}
Update-sync for a Memory
Updates the Memory with given TMX file. Header parameters should be passed as JSON object with the header field LILT-API. The number of segments returned by the from_time, to_time, when parameters and the number of segments in the TMX file need to be identical.
Example CURL command:
curl -X PUT https://lilt.com/2/memories/sync?key=API_KEY \
--header "LILT-API: {\"name\": \"my_memory.tmx\", \"id\": 42, \"from_time\": 1491048000, \"to_time\": 1491049800, "when": "Updated"}" \
--header "Content-Type: application/octet-stream" \
--data-binary @my_memory.tmx
The file contents to be uploaded. The entire PUT body will be treated as the file.
A unique Memory identifier.
Unix time stamp (epoch, in seconds) of the start of the Memory section.
Unix time stamp (epoch, in seconds) of the end of the Memory section.
The date field on which retrieved segments match from/to time stamps: Created, Updated, Deleted.
Request Example
"string"
OK
Response Example (200 OK)
{
"id": "integer",
"num_updates": "integer"
}
Delete-sync for a Memory
Deletes segments in the Memory matching the from_time, to_time and when parameters.
Example CURL command:
curl -X DELETE https://lilt.com/2/memories/sync?key=API_KEY&id=42&from_time=1491048000&to_time=1491049800&when=Created
A unique Memory identifier.
Unix time stamp (epoch, in seconds) of the start of the Memory section.
Unix time stamp (epoch, in seconds) of the end of the Memory section.
The date field on which retrieved segments match from/to time stamps: Created, Updated, Deleted.
OK
Response Example (200 OK)
{
"id": "integer",
"num_updates": "integer"
}
Projects
Create a Project
Create a Project. A Project is a collection of Documents.
A Project is associated with exactly one Memory.
Projects appear in the dashboard of the web app.
The Project resource to create.
- name: string
-
A name for the Project.
- memory_id: integer
-
The Memory to associate with this new Project.
Request Example
{
"name": "My new project",
"memory_id": 1234
}
OK
Response Example (200 OK)
{
"id": 448,
"memory_id": 1234,
"name": "My new project",
"srclang": "en",
"trglang": "fr",
"documents": []
}
Retrieve a Project
Retrieve a project.
A project id.
An ISO639-1 language code.
An ISO639-1 language code.
OK
Response Example (200 OK)
[
{
"id": 448,
"memory_id": 1234,
"name": "My new project",
"srclang": "en",
"trglang": "fr",
"documents": []
}
]
Update a Project
Update a Project.
The Project resource to update.
- id: integer
-
A unique Project identifier.
- name: string
-
The Project name.
Request Example
{
"id": 1234,
"name": "Walker Percy Essays"
}
OK
Response Example (200 OK)
{
"id": 1234,
"name": "Walker Percy Essays"
}
Delete a Project
Delete a Project.
A unique Project identifier.
OK
Retrieve Project status
Retrieve the status of a project.
A project id.
OK
Response Example (200 OK)
{
"id": 21902,
"num_source_words": 6039,
"num_words_confirmed": 151,
"num_words_reviewed": "integer",
"time_elapsed": 2980000,
"time_elapsed_review": "integer",
"resources": [
{
"email": "support@lilt.com",
"num_words_confirmed": 151,
"time_elapsed": 1172000,
"num_exact_match_words": 52,
"num_reviewed_words": "integer",
"time_elapsed_review": "integer"
}
]
}
QA
Perform QA check
Perform QA checks on a target string. Optionally, you can specify a source string for additional bilingual checks, e.g. number consistency.
A target string to be checked.
An ISO639-1 language code.
An optional source string.
An ISO639-1 language code.
OK
Response Example (200 OK)
{
"matches": [
{
"context": {
"length": 7,
"offset": 19,
"text": "This segment has a speling mistake"
},
"length": 7,
"message": "Possible spelling mistake found",
"offset": 19,
"replacements": [
{
"value": "spelling"
},
{
"value": "spewing"
},
{
"value": "spieling"
}
],
"rule": {
"category": {
"id": "TYPOS",
"name": "Possible Typo"
},
"description": "Possible spelling mistake",
"id": "MORFOLOGIK_RULE_EN_US",
"issueType": "misspelling",
"subId": "string",
"urls": []
},
"shortMessage": "Spelling mistake"
}
]
}
Segments
Create a Segment
Create a Segment and add it to a Memory. A Segment is a source/target pair that is used to train the machine translation system and populate the translation memory.
The maximum source length is 5,000 characters.
The Segment resource to create.
- memory_id: integer
-
A unique Memory identifier.
- source: string
-
The source string.
- target: string
-
The target string.
Request Example
{
"memory_id": 10641,
"source": "Code zur Fehleranalyse einschalten",
"target": "Enable debugging code"
}
OK
Response Example (200 OK)
{
"id": 84480010,
"created_at": "2017-05-15T13:54:02.000Z",
"document_id": null,
"memory_id": 10641,
"srclang": "en",
"target": "Enable debugging code",
"trglang": "de",
"is_confirmed": true,
"is_reviewed": true
}
Retrieve a Segment
Retrieve a Segment.
A unique Segment identifier.
OK
Response Example (200 OK)
{
"id": 84480010,
"created_at": "2017-05-15T13:54:02.000Z",
"document_id": null,
"memory_id": 10641,
"srclang": "en",
"target": "Enable debugging code",
"trglang": "de",
"is_confirmed": true,
"is_reviewed": true
}
Update a Segment
Update a Segment. The Memory will be updated with the new target string.
The Segment resource to update.
- id: integer
-
A unique Segment identifier.
- target: string
-
The target string.
Request Example
{
"id": 84480010,
"target": "Enable debug code"
}
OK
Response Example (200 OK)
{
"id": 84480010,
"created_at": "2017-05-15T13:54:02.000Z",
"document_id": null,
"memory_id": 10641,
"srclang": "en",
"target": "Enable debugging code",
"trglang": "de",
"is_confirmed": true,
"is_reviewed": true
}
Delete a Segment
Delete a Segment.
A unique Segment identifier.
OK
Translate
Translate a segment
Translate a source string. Setting the rich parameter to true will change the response format to include additional information about each translation including a model score, word alignments, and formatting information. The rich format can be seen in the example response on this page. The maximum source length is 1,500 characters.
A unique Memory identifier.
The source text to be translated.
A source hash code.
A target prefix.
Return top n translations.
Returns rich translation information (e.g., with word alignments).
Include translation memory matches.
OK
Response Example (200 OK)
{
"untokenizedSource": "Authentication not required.",
"tokenizedSource": "Authentication not required .",
"sourceDelimiters": [
"",
" ",
" ",
"",
""
],
"translation": [
[
{
"score": 3.4936864e-8,
"align": "0-0 1-1 2-2 3-3",
"targetDelimiters": [
"",
" ",
" ",
"",
""
],
"targetWords": [
"Authentifizierung",
"nicht",
"erforderlich",
"."
],
"target": "Authentifizierung nicht erforderlich .",
"targetWithTags": "Authentifizierung nicht erforderlich.",
"isTMMatch": false,
"provenance": "0 0 0 0"
}
]
]
}
Register a segment
Register a source string for interactive translation. The source_hash value that is returned by this request is required by the prefix parameter for the translation endpoint.
The maximum source length is 1,500 characters.
A source string to be registered.
An ISO639-1 language code.
An ISO639-1 language code.
OK
Response Example (200 OK)
{
"source_hash": "integer",
"num_words": "integer"
}
Schema Definitions
Project: object
A Project is a collection of zero or more Documents. It is specific to a language pair, and is associated with exactly one Memory for that language pair. The Memory association cannot be changed after the Project is created.
- id: integer
-
A unique number identifying the Project.
- memory_id: integer
-
A unique number identifying the associated Memory.
- srclang: string
-
An ISO639-1 language identifier.
- trglang: string
-
An ISO639-1 language identifier.
- name: string
-
A name for the project.
- documents: integer[]
-
A list of Document identifiers.
Example
{
"id": 448,
"memory_id": 1234,
"name": "My new project",
"srclang": "en",
"trglang": "fr",
"documents": []
}
Document: object
A Document is a collection of zero or more Segments.
- id: integer
-
A unique number identifying the Document.
- srclang: string
-
An ISO639-1 language identifier.
- trglang: string
-
An ISO639-1 language identifier.
- name: string
-
The document name.
- segments: integer[]
-
A list of Segment ids.
Example
{
"id": 46530,
"srclang": "en",
"trglang": "de",
"name": "Introduction.xliff",
"segments": [
"integer"
]
}
LexiconEntry: object
An Lexicon entry for a source term or phrase.
- translations: object[]
-
A list of translations for the query term.
- examples: object[]
-
A list of concordance examples for the query term.
Example
{
"translations": [],
"examples": [
[
{
"source": "Waschmaschine",
"sourceDelimiters": [
"",
""
],
"target": [
"washing",
"machine"
],
"targetDelimiters": [
"",
" ",
""
],
"sourceSpan": {
"start": 0,
"length": 1
},
"targetSpan": {
"start": 1,
"length": 1
},
"similarity": 0,
"memoryId": 10641
}
]
]
}
Memory: object
A Memory is a collection of parallel (source/target) segments from which a MT/TM model is trained. When a translator confirms a segment in the Interface, a parallel segment is added to the Memory. Parallel segments from existing translation memories and bitexts can also be added to the Memory via the API.
- id: integer
-
A unique number identifying the Memory.
- srclang: string
-
An ISO639-1 language identifier.
- trglang: string
-
An ISO639-1 language identifier.
- name: string
-
A name for the Memory.
- num_segments: integer
-
The number of confirmed Segments incorporated into this Memory.
Example
{
"id": 7245,
"name": "automotive",
"num_segments": 890,
"srclang": "en",
"trglang": "fr"
}
Translation: object
A machine translation (MT) or a translation memory (TM) match of a source segment.
- target: string
-
The target string.
- targetWithTags: string
-
The target string with source tags projected into the target.
- align: string
-
"MT only: A whitespace delimited list of source-target alignment indices."
- provenance: string
-
Positive values indicate that the word is from the Memory, with contiguous identical entries (e.g., 2 2) indicating phrase matches. Negative contiguous values indicate entries from the Lexicon. 0 indicates a word from the background data.
- score: number
-
The score of the translation.
- isTMMatch: boolean
-
TM only: If true, indicates an exact translation memory match.
- targetDelimiters: string[]
-
A format string that indicates, for each word, if the word should be preceded by a space.
- targetWords: string[]
-
A list of target words corresponding with the same dimension as
targetDelimiters. The target string can be constructed by prefixing each word with its corresponding value intargetDelimiters, and then concatenating the words.
Example
[
{
"score": 3.4936864e-8,
"align": "0-0 1-1 2-2 3-3",
"targetDelimiters": [
"",
" ",
" ",
"",
""
],
"targetWords": [
"Authentifizierung",
"nicht",
"erforderlich",
"."
],
"target": "Authentifizierung nicht erforderlich .",
"targetWithTags": "Authentifizierung nicht erforderlich.",
"isTMMatch": false,
"provenance": "0 0 0 0"
}
]
TranslationList: object
An ranked list of translations and associated metadata.
- untokenizedSource: string
-
The untokenized source segment. Punctuation has not been separated from words.
- tokenizedSource: string
-
The tokenized source segment. Punctuation has been separated from words.
- sourceDelimiters: string[]
-
A format string that indicates, for each word, if the word should be preceded by a space.
- translation: object[]
-
One or more Translation objects.
Example
{
"untokenizedSource": "Authentication not required.",
"tokenizedSource": "Authentication not required .",
"sourceDelimiters": [
"",
" ",
" ",
"",
""
],
"translation": [
[
{
"score": 3.4936864e-8,
"align": "0-0 1-1 2-2 3-3",
"targetDelimiters": [
"",
" ",
" ",
"",
""
],
"targetWords": [
"Authentifizierung",
"nicht",
"erforderlich",
"."
],
"target": "Authentifizierung nicht erforderlich .",
"targetWithTags": "Authentifizierung nicht erforderlich.",
"isTMMatch": false,
"provenance": "0 0 0 0"
}
]
]
}
Segment: object
A Segment is a source string and, optionally, its translation. A Segment can be associated with both a Memory and a Document. The Segment object contains additional metadata about the source and target strings.
- id: integer
-
A unique number identifying the Segment.
- created_at: string (date)
-
An ISO8601 date.
- document_id: integer
-
A unique Document identifier.
- memory_id: integer
-
The Memory with which this Segment is associated.
- srclang: string
-
An ISO639-1 language code.
- target: string
-
The target string.
- trglang: string
-
An ISO639-1 language code.
- is_confirmed: boolean
-
The confirmation status.
- is_reviewed: boolean
-
The review status.
Example
{
"id": 84480010,
"created_at": "2017-05-15T13:54:02.000Z",
"document_id": null,
"memory_id": 10641,
"srclang": "en",
"target": "Enable debugging code",
"trglang": "de",
"is_confirmed": true,
"is_reviewed": true
}
TranslationMemoryEntry: object
A translation memory entry.
- created_at: string (date)
-
An ISO8601 date.
- memory_id: integer
-
The Memory with which this Segment is associated.
- source: string
-
The source string.
- created_by: object
-
Identifier of the user who created this entry.
Example
{
"created_at": "string (date)",
"memory_id": "integer",
"source": "string",
"created_by": {
"email": "string",
"name": "string"
}
}
QARuleMatches: object
QA rules describing the errors in the text.
- matches: object[]
Example
{
"matches": [
{
"context": {
"length": 7,
"offset": 19,
"text": "This segment has a speling mistake"
},
"length": 7,
"message": "Possible spelling mistake found",
"offset": 19,
"replacements": [
{
"value": "spelling"
},
{
"value": "spewing"
},
{
"value": "spieling"
}
],
"rule": {
"category": {
"id": "TYPOS",
"name": "Possible Typo"
},
"description": "Possible spelling mistake",
"id": "MORFOLOGIK_RULE_EN_US",
"issueType": "misspelling",
"subId": "string",
"urls": []
},
"shortMessage": "Spelling mistake"
}
]
}