Offensive language detection for Turkish text.
Send a Turkish comment, message or review as it was written. Detectivision answers with each flag it raised, such as an insult or hate speech, and how sure it is. English works the same way, in the same call.
Five flags for offensive Turkish
Every account has these text flags. Ask for all of them, or only those your rules need.
Insults and personal attacks on someone.
Hate against a group of people for who they are.
Racism · ırkçılık
Racist or ethnic hate speech.
Sexism · cinsiyetçilik
Sexist statements.
Sexual language, or sexual harassment of someone.
Check a Turkish comment
curl https://api.detectivision.ai/api/v1/moderation \
-H "Api-Key: $DETECTIVISION_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"type": "text_moderation",
"input": "Sen salak birisin, kimse seni istemiyor.",
"mode": "detect",
"flags": ["insult", "hate", "racism", "sexism", "sexual"]
}'{
"id": 9105,
"type": "text_moderation",
"mode": "detect",
"status": "completed",
"flagged": true,
"flags": [{ "flag": "insult", "confidence": 0.94 }],
"checked": ["hate", "insult", "racism", "sexism", "sexual"],
"unchecked": [],
"tookMs": 610,
"usage": { "charged": 2, "remaining": 9998 }
}The text is checked exactly as you send it, Turkish characters and all: nothing is trimmed or translated first.
Send it as written
No need to translate or clean up a text first. Send it with its Turkish characters, up to 10,000 characters.
Act per flag
Remove hate and racism at once, hold insults for a moderator, or only warn. Each flag is its own decision.
12 languages, one call
There is no language to set: texts in English, Spanish, Portuguese, French, German, Italian, Turkish, Arabic, Russian, Hindi, Chinese and Japanese go to the same endpoint, so a community that writes in several of them needs one integration.
Forums, comment sections and social apps.
Chat, player names and clan names.
Reviews and messages between buyers and sellers.
Why English-first tools miss Turkish and Arabic abuse, and how to test an API on them.
Perspective lists no Turkish for any attribute, and it stops after 31 December 2026.
Frequently asked
Do I need to say which language a text is in?
No. There is no language setting: send the text, in Turkish or in English.
Does it catch Turkish swear words?
Yes, on the Basic plan: it has a profanity flag for swear words and vulgar slang, even when they are not aimed at anyone. Swearing aimed at someone also raises insult, and words such as salak or aptal count as insults, not profanity. GET /api/v1/flags lists your account’s own flags.
Which other languages work?
English, Spanish, Portuguese, French, German, Italian, Turkish, Arabic, Russian, Hindi, Chinese and Japanese. Text in other languages is checked, but the results are not reliable.
What does a check cost?
A text costs 1 credit per started 2,500 characters when the answer comes by webhook, and twice that when it comes at once. Asking for fewer flags never costs more.
Is there a free trial?
Yes. The plans, and the free trial, are on the pricing page.