POST /detect-language

Summary

Detects the natural language of a given text input and returns the identified language code.

Intent

Enables applications to automatically identify the language of user-submitted text without requiring the caller to know the language in advance, supporting multilingual workflows such as routing, translation, and content classification.

Parameters

Name	Type	Required	Description
text	string	Yes	The text content whose language should be detected. Should be a meaningful sample of text; very short or ambiguous strings may reduce detection accuracy.

Request example

import requests

response = requests.post(
    "https://api.example.com/detect-language",
    json={
        "text": "Bonjour, comment allez-vous aujourd'hui?"
    }
)

result = response.json()
print(result["language"])  # e.g., "fr"
print(result["confidence"])  # e.g., 0.99

Response example

{
  "language": "fr",
  "language_name": "French",
  "confidence": 0.99,
  "alternatives": [
    {
      "language": "it",
      "confidence": 0.005
    },
    {
      "language": "es",
      "confidence": 0.003
    }
  ]
}

Error cases

• 400 — when the request body is missing the required ‘text’ field or the field is empty
• 422 — when the text is too short or contains only symbols/numbers and language cannot be reliably determined
• 500 — when an internal error occurs during language detection processing

Gotchas

• Very short strings (fewer than ~10 characters) or strings composed entirely of numbers, punctuation, or emojis may return low confidence scores or fail detection entirely.
• Mixed-language text (code-switching) will typically return only the dominant language detected, not all languages present.
• The endpoint does not require authentication, but may still be subject to rate limiting based on IP address or API key if one is provided.
• Language codes returned follow ISO 639-1 (two-letter) format by default; verify this matches your downstream system’s expectations.
• Confidence scores close to each other across multiple alternatives indicate ambiguous input — treat results with low confidence as unreliable.
• Sending personally identifiable information (PII) in the text field may have privacy implications depending on how the service processes and logs requests.