Language Detector
Identify the language of any text with a confidence score across 50+ languages
/v1/text/detect-language
How to Use
1. Send a POST request to `/v1/text/detect-language` with the `text` field. 2. Read `language_code` (ISO 639-1, e.g. `"fr"`) and `language_name` (e.g. `"French"`). 3. Use `confidence` to decide whether to trust the result — scores above 0.85 are highly reliable. 4. For very short texts (< 20 words), treat results with confidence < 0.7 as uncertain.
About This Tool
The Language Detection endpoint identifies the language of any text snippet and returns an ISO 639-1 language code along with a confidence score and the full language name.
Detection is based on character n-gram frequency analysis — a lightweight, fast approach that works well for texts of 20 words or more across 50+ languages including all major European languages, Chinese, Japanese, Korean, Arabic, and Russian.
Use Cases
Multi-language content routing
Route user-submitted content to the correct language processing pipeline
Automatic translation trigger
Detect non-English content and trigger a translation workflow
Analytics segmentation
Segment user feedback by language for localised analysis
Form validation
Verify that content is in the expected language before processing