MTQE API (Machine Translation Quality Estimation)

The MTQE API allows evaluating translation quality automatically without the need for a human reference. It provides a score indicating the precision and fluency of the translated text relative to its source text.

To use the API, an apikey provided by Pangeanic at the time of sale is required.

Supported Languages (MTQE)

The following table details the available language pairs, including their names and the standardized encoding for API use.

Notes:

• Two-letter codes are used for most languages.
• Usage Note: Codes in API calls must always be sent in lowercase.

Available languages (click to expand)

Source Language	Target Language	Source	Target
English	Czech	en	cs
English	German	en	de
English	Spanish	en	es
English	Estonian	en	et
English	Finnish	en	fi
English	French	en	fr
English	Gujarati	en	gu
English	Hindi	en	hi
English	Croatian	en	hr
English	Hungarian	en	hu
English	Icelandic	en	is
English	Japanese	en	ja
English	Kazakh	en	kk
English	Khmer	en	km
English	Lithuanian	en	lt
English	Latvian	en	lv
English	Polish	en	pl
English	Romanian	en	ro
English	Russian	en	ru
English	Tamil	en	ta
English	Turkish	en	tr
English	Ukrainian	en	uk
English	Chinese	en	zh
Czech	English	cs	en
German	English	de	en
Finnish	English	fi	en
Hebrew	English	he	en
Icelandic	English	is	en
Inuktitut	English	iu	en
Kazakh	English	kk	en
Livonian	English	liv	en
Pashto	English	ps	en
Russian	English	ru	en
Turkish	English	tr	en
Chinese	English	zh	en
German	French	de	fr
French	German	fr	de
Czech	Ukrainian	cs	uk
Ukrainian	Czech	uk	cs
Japanese	Chinese	ja	zh
Upper Sorbian	German	hsb	de
German	Upper Sorbian	de	hsb

---

POST /score

Calculates the quality score for one or several segment pairs. The value will be between 0 and 1.

URL: https://api.pangeanic.com/service/mtqe/v1/score

Request Parameters

Parameter	Type	Location	Description
p-api-key	string	Header	Your API access key.
source_language	string	Body	Source language code (e.g., "en"). See supported languages
target_language	string	Body	Target language code (e.g., "es"). See supported languages
pairs	array	Body	List of objects with source and target.

PythoncURLOutput (JSON)

import requests

url = "https://api.pangeanic.com/service/mtqe/v1/score"
headers = {
    "accept": "application/json",
    "p-api-key": "your-api-key-here",
    "Content-Type": "application/json"
}

payload = {
    "source_language": "en",
    "target_language": "es",
    "pairs": [
        {
            "source": "The quick brown fox jumps over the lazy dog.",
            "target": "El veloz zorro marrón salta sobre el perro perezoso."
        },
        {
            "source": "Never say never.",
            "target": "Nunca digas nunca."
        }
    ]
}

res = requests.post(url, json=payload, headers=headers)
print(res.json())

curl -X 'POST' \
  'https://api.pangeanic.com/service/mtqe/v1/score' \
  -H 'accept: application/json' \
  -H 'p-api-key: your-api-key-here' \
  -H 'Content-Type: application/json' \
  -d '{
    "source_language": "en",
    "target_language": "es",
    "pairs": [
      {
        "source": "The quick brown fox jumps over the lazy dog.",
        "target": "El veloz zorro marrón salta sobre el perro perezoso."
      },
      {
        "source": "Never say never.",
        "target": "Nunca digas nunca."
      }
    ]
  }'

{
  "pairs": [
    {
      "score": 0.9854,
      "source": "The quick brown fox jumps over the lazy dog.",
      "target": "El veloz zorro marrón salta sobre el perro perezoso."
    },
    {
      "score": 0.9998,
      "source": "Never say never.",
      "target": "Nunca digas nunca."
    }
  ]
}

Score Interpretation

Score Range	Estimated Quality	Recommended Action
0.90 - 1.00	Excellent	Direct publication or minimal review.
0.70 - 0.89	Good	Light human review (light MTPE).
0.40 - 0.69	Medium / Low	Requires mandatory deep post-editing.
< 0.40	Very Low	Invalid translation or incoherent content.

Disclaimer

This interpretation is solely a suggestion based on statistical correlation with human perception. The final decision on quality will always depend on the degree of strictness and the specific quality requirements (KPIs) required for each translation project.

Common Errors

• 400 (Bad Request): The JSON body is invalid or mandatory fields are missing.
• 401 (Unauthorized): The p-api-key provided in the header is incorrect or has expired.
• 422 (Unprocessable Entity): The specified language is not supported by the current MTQE engine.