Language Detection as a Service

Detecting 173 languages and profanity in real-time. Powered by humans and AI.

Start free

and stay there
as long as you want

100 detections / day

no credit card required

Get API Key

Pay as you grow

$ / month

No subscription! You get charged once per month based on the maximum daily detections in the prior month.

Get API Key

need more?

more detections, dedicated instance, SLA, fixed price, premium support, ...

get in touch

Getting started couldn't be easier

There is a library for you:

curl https://api.linguin.ai/v2/detect/language \
-H "Authorization: Bearer <API_KEY>" \
-d 'q=What language is this?'
# => {"results":[{"lang":"en","confidence":1.0}]}

curl https://api.linguin.ai/v2/detect/profanity (...)
# => {"score":0.9981}

response = Linguin.detect_language("What language is this?")
response.results
=> [{ lang: "en", confidence: 1.0 }]

Linguin.detect_profanity("YOU MORON!").score
=> 0.9981

response = linguin.detect_language("What language is this?")
response.result
# >> { 'results': [{ 'lang': 'en', confidence: 1.0 }, ...] }

linguin.detect_profanity("YOU MORON!").result
# >> { 'score': 0.9981 }

final Languages detected = linguin.detect("What language is this?");
System.out.println("Most probable language: " + detected.bestMatch().code());
// "Most probable language: en"

API Documentation

When will I get charged for what?

You can sign up without a credit card. You will never be charged anything, but you are limited to 100 detections per day.

If you upgrade your account by providing a credit card, you will be charged once per month based on the busiest days (the day with the most detections) within the past month. Pricing starts at $2.50 for up to 1.000 detections + $0.15 for each additional 1.000 detections.

If you stay below 101 detections the entire month, you won't be charged anything.

For full control over your spending, you can set a daily detection limit in your account.

How does the detection work?

Linguin uses a combination of several different approaches to maximize accuracy and speed. To determine the best way to detect the language Linguin performs some static analysis before feeding text into the most fitting detection AI.

For profanity detection a set of trained machine-learning models is used.

We accept and appreciate user input via https://linguin.ai/was/wrong to improve detection accuracy even further.

What happens to my data?

We do not store any text submitted via our API. The text is permanently deleted - including from any log files - as soon as the detection results are returned.

Our company and infrastructure is located in Germany. Therefore, all GDPR rules apply to us and are followed. Data is generally only stored and processed when necessary or required by law. We do not sell your data to advertisers or anyone else.